HansDeLeenheer

The Company Library 2.0?

2012-01-06T13:54:00.000+01:00

As it should be at a self-respecting tech company, we have a library at every floor. If somebody wants a new book, we just ask and it will be ordered. You do need to give some kind of argumentation why but I haven't been rejected one yet. Maybe my arguments are good enough or I order the right books :-)

These are some I ordered through the last months:
* VMware vSphere Design - Forbes Guthrie, Scott Lowe, Maish-Saidel-Keesing
* VMware vSphere 4.1 HA and DRS Technical deepdive - Duncan Epping, Frank Denneman
* VMware vSphere 5 Clustering Technical Deepdive - Duncan Epping, Frank Denneman
* Mastering VMware vSphere 5 - Scott Lowe
* Cloud and Virtual Data Storage Networking - Greg Schulz

What's next? I want to read these books off course and I want to share with other colleagues so we can discuss the content. But where do you want to read them? Fank and Duncan write small enough but if you want to take 2 Scotts and a Greg you'll need a backpack for your books alone. So I buy an iPad. Mmmm ... issue ... these paperbaks don't fit into my iPad. So I bought them again through my own Amazon account. You guess where I am going here: my wife doesn't like my job so much anymore.

And then THE solution came through Lendle.me, an online company that allows you to lend out Kindle books you bought. you can give in the books you own and others can lend them. Not all books are lendable. This has to be enabled by the producers (or writers).

Example:

I truly loved the idea. Put in all my books and even payed my € 19 te become a "patron". One of the things a patron can do is create bookclubs. So I did. I created a bookclub for me and my colleagues and invited some of them to be testing dummies.

So lets go, share this sh*t and read it! IT DIDN'T WORK. I'll just give you the screenshot and leave it with that.

I know the restrictions are on the Amazon side of the story but nevertheless it is useless for us. So I beg the community out there: give me a new solution please that works for us.

=======================================

Today I saw that Duncan mentioned borrowing his book on his own website. Read this post how to. You'll need an Kindle Prime subscription.

RDP on View4.5 versus PCoIP on View5

2011-12-19T17:28:00.001+01:00

CASE:

VDI environment at customer site with an average user base of 300 clients.

Issues: performance (duuuh!)

Consultancy:

January 2011: dear customer, buy more spindles please!

January 2011: dear customer, can we please change from view4.5 to 4.6 so we can use PCoIP?

The customer:

January 2011: $$$$$ .... NO

February 2011: $$$$$ .... NO

March 2011: ...

December 2011: please solve our performance issues before the end of the year!

Consultancy:

(thoughts) ARE YOU KIDDING ME?

(answer) yes mr customer, we will give you our own hardware and prove you wrong. On the other hand, we are allready one year further so we offer a new solution to your issues: 2 new blades wth FusionIO cards and VIEW5.

============================================

Results of POC (Proof-Of-Concept).

VIEW5 setup on vSphere5 with Win7 linked clone image and no more tweaks then necessary. For the VIEW we used the VIEW 5 evaluators guide (download here). For the Win7 image we used the VMware View Optimization Guide for Windows 7 (download here)

Test 1) go to www.wechoosethemoon.org and launch the Apollo 11, just 40 years later. you can view the result of both sessions below.

Session 1: VIEW4.5 with RDP
Session 2: VIEW5 with PCoIP

Test 2) the test above is still user experience so I did the test twice and added a powerpoint show of over 5Mb. I captured these sessions with Wireshark (filter: host ip) and this is the amazing result!

Disclaimer: this is "only" a proof of concept on really old hardware (10 VDIs on 2 x HP BL460G1 with 12Gb RAM each) so it only shows the difference for this configuration and is not a benchmark to be used. There are too many different parameters that are changed. The result here is a sum of all enhancements:
* Client Operating system from Vista to W7
* hypervisor from vSphere4.1 to vSphere5
* VIEW 4.5 to VIEW5
* RDP to PCoIP

Lessons learned: should have done this a year ago ...

#ChuckNorris is the reason why Waldo is hiding

2011-12-14T22:46:00.000+01:00

1. #ChuckNorris invented black. In fact, he invented the entire spectrum of visible light. Except pink. Tom Cruise invented pink.

2. #ChuckNorris has already been to mars, that's why there are no signs of life there.

3. There is No theory of evolution, just a list of creatures #ChuckNorris allows to live

4. #ChuckNorris is the reason why Waldo is hiding.

5. #ChuckNorris & Superman had an armwrestling bet, and agreed the loser would wear his underpants on his trousers.

6. The grass is always greener on the other side, unless #ChuckNorris has been there. In that case the grass is most likely soaked in blood and tears.

7. When God said "let there be light", #ChuckNorris said "say please"

8. They once made a #ChuckNorris toilet paper, but it wouldn't take shit from anybody.

9. #ChuckNorris threw a grenade & killed 50 people. Then the grenade exploded

10. A blind man once stepped on #ChuckNorris ' shoe. Chuck replied, "Don't you know who I am? I'm #ChuckNorris!" The mere mention of his name cured this man blindness. Sadly the first, last, and only thing this man ever saw, was a fatal roundhouse delivered by #ChuckNorris.

11. #ChuckNorris doesn't flush toilets He scares the shit out of them

12. The Four Horsemen of the Apocalypse actually live in #ChuckNorris's nutsack.

13. the only reason we have a moon is because #ChuckNorris gave the sun a curfew

14. When observing a #ChuckNorris roundhouse kick in slow motion, one finds that #ChuckNorris actually rapes his victim in the ass, smokes a cigarette with Dennis Leary, and then roundhouse kicks them in the face.

15. ChuckNiorris once urinated into a semi trucks gas tank. That truck now is known as Optimus Prime

16. If you have five dollars and #ChuckNorris has five dollars, #ChuckNorris has more money than you.

17. #ChuckNorris doesnt need Twitter, you cant follow him

18. #ChuckNorris sent Jesus a birthday card on December 25th and it wasn't Jesus’ birthday. Jesus was too scared to correct #ChuckNorris and to this day December 25th is known as Jesus' birthday.

19. If the boogeyman goes to sleep he'll check if #ChuckNorris is in the closet

20. When #ChuckNorris had surgery, the anesthesia was applied to the doctors.

21. #ChuckNorris beats a Royal Flush with one pair

22. #ChuckNorris once broke the land speed record on a bicycle that was missing its chain and the back tire.

23. When Alexander Bell invented the telephone he had 3 missed calls from #ChuckNorris

24. Multiple people have died from #ChuckNorris giving them the finger.

25. The only poster #ChuckNorris has on his bedroom wall is a mirror

26. Pee Wee Herman got arrested for masturbating in public. The same day, #ChuckNorris got an award for masturbating in public.

27. #ChuckNorris counted to infinity - twice

28. Once a grizzly bear threatened to eat #ChuckNorris. Chuck showed the bear his fist and the bear proceeded to eat himself, because it would be the less painful way to die.

29. #ChuckNorris isn't superhuman,if he was, he'd come to my house & smash my face into my keyboa4egdvc8 ohuilj4g evda zcg 9ob jmevda z :63e2wt#

30. If #ChuckNorris is late, time better slow the fuck down

31. #ChuckNorris doesnt do push-ups, he pushes the earth down

32. #ChuckNorris can touch MC Hammer.

33. They once showed a #ChuckNorris movie in 3D.... There were no survivors

34. #ChuckNorris ordered a Big Mac at Burger King, and got one.

35. only #ChuckNorris can slam revolving doors

36. #ChuckNorris can capitalize numbers

37. #ChuckNorris can draw a perfect circle... With a ruler

38. #ChuckNorris always has sex on the first date. Always.

39. #ChuckNorris does not love Raymond

40. A Handicap parking sign does not signify that this spot is for handicapped people. It is actually in fact a warning, that the spot belongs to #ChuckNorris and that you will be handicapped if you park there.

41. #ChuckNorris won the World Series of Poker using Pokemon cards

42. There is no such thing as tornados. #ChuckNorris just hates trailer parks.

43. #ChuckNorris doesn't worry about changing his clock twice a year for daylight savings time. The sun rises and sets when Chuck tells it to.

44. TimTebow wears #ChuckNorris Pajamas

45. #ChuckNorris only masturbates to pictures of #ChuckNorris .

46. #ChuckNorris found out his daughter lost her virginity... He found it and got it back

47. #ChuckNorris was the fourth Wiseman. He brought baby Jesus the gift of “beard”. Jesus wore it proudly to his dying day. The other Wisemen, jealous of Jesus’ obvious gift favoritism, used their combined influence to have Chuck omitted from the Bible. Shortly after all three died of roundhouse kick related deaths.

48. Brett Favre retires every year ‘cause he's afraid #ChuckNorris might try out for NFL

49. Although it is not common knowledge, there are actually three sides to the Force: the light side, the dark side, and #ChuckNorris .

50. #ChuckNorris is always on top when having sex, because #ChuckNorris never fucks up

51. A duck’s quack does not echo. #ChuckNorris is solely responsible for this phenomenon. When asked why he will simply stare at you, grimly.

52. #ChuckNorris doesn’t believe in Germany.

53. #ChuckNorris climbed on the top of soundwall

54. #ChuckNorris slept through 9/11. His bedroom was on the 73rd floor

55. The reason WTC fell was because #ChuckNorris thought they were his birthday candles and he blew them out.

56. When has #ChuckNorris lost his virginity? Never! #ChuckNorris never loses

57. The Black-Eyed Peas were simply known as The Peas until they crossed #ChuckNorris

58. Rosa Parks refused to get out of her seat becuz she was saving it for #ChuckNorris

59. If #ChuckNorris were a woman, he wouldn't have periods. He would have an exclamation point

60. The saddest moment for a child is not wen they learn Santa isnt real, its wen they learn #ChuckNorris is

61. #ChuckNorris once had a heart attack. His heart lost

HP Cumulonimbus - Verticaly developed Cloud

2011-12-06T20:30:00.001+01:00

I was looking for an image to accompany my Cloud blog and came across the "Cumulonimbus". Together with the following text I couldn't resist using it as the starting punchline: "Cumulonimbus clouds (Cb) are much larger and more vertically developed than fair weather cumulus..."

So what do I want to tell you? We have seen and heard about cloud enough. More then enough I guess. Every time another vendor starts another session on clouds we are already running for coffee (or beers if its the last session of the day). I found out why: they all are telling us what we should do in the future, what is the path we are going to follow together without knowing what the endpoint will be? Have trust in the vendors insights ... blablabla

Last week I heard a different story at HP Discover. HP showed us where they are now. What can they offer you now. Here are the products, this is how it works and here is how we are going to help you.

The Matrix:
Until recently we talked about HP BladeSytem Matrix. Rome wasn't build in a day and so wasn't the Cloud. The Cloudsystem Matrix now goes way beyond that. Instead of discussing hardware and features we need to start discussing services. This shows for example in the implementation of the "Storage Provisioning Manager". Here the storage architect can set-up storage templates (array - tier - raid - TP - size - ...) give it a chargeback value and set permissions on those templates. The cloud admin can deploy those templates without any storage knowledge.

Another new item in the CloudSystem Matrix is the builtin "burst" capabilities provided by Savvis. Essentially this means your private cloud has endless resources. The big question here remains how to get your data back when you don't need the burst anymore. This is still a bit unanswered.

A third item HP integraded is a cooperation with "Alcatel-Lucent" that your burst mechanism would choose the right datacentre to burst in according to best availability such as bandwith and latency. This is the part that still has the most questions for me but the idea of it strokes with the entire concept.

Implementation & Best Practices:

To make sure everything is integraded according to best practises HP will provide us with Reference Architectures, Strategy- Consultancy- and Implementation- services and they will continu to invest in labs with other strategic partners (vmware, Microsoft, McAfee, ...).

Something I heared before from other vendors is that they do not want to sell people as a product. This means that they still want to keep on focussing on the channel (keep in mind that this is the cheapest sales force for a vendor). The only down side for the current resellers is that the focus will change from hardware box movement to service provider. I quote one speaker: "we can only hope it will still be the same people" wich means: lead, follow or get out of the way :-)

Will they help us getting there? YES! HP just added 3 new certifications to its tracks: Cloud Architect ASE, Cloud Integrator ASE & Cloud Integrator Master ASE. Last year I noted at an HP conference that they appologized for giving us education on yesterdays products and that they would get this straight this year. So what else did they do? Well, there are 5 new learning solution partners, there are press/self study books that are coming out very soon and last but not least HP is going to school! "HP institute" will go into beta starting this january in several highschools. And in contrast to what other vendors did before (cfr Cisco) HP will provide them with remote labs so that the schools do not need to invest in all latest technology for lab environments. I personally think this is huge! We might finally get proper educated students in the IT market.

Other Services:

There is one other Cloud thing I want to share. During the HP Discover conference we were laughing at the printing people. Who the hell wants to blog about printing. What can your printer do? It can print colors! #LOL. And still HP managed to get my attention. In all new network printers HP added a feature called e-print. This feature registers the printer in a private, public or enterprise section of the HP printing cloud and enables you to print to that printer from any mobile device without the need of installing drivers for each printer. You simply click "print" at your device and you can choose from all printers in your neighbourhood (via gps or a simple search). Think about printing your boarding pass you got by email. Once at the printer, you get the document from the cloud with the passcode you gave it. You can also create a private Enterprise e-print network and print an internal note at the printer in HongKong for example. The consumer edition is free to use! Only downside: there is no "proxy box" available to put in front of existing printers so at this stage it comes down to a nice feature on NEW printers.

To summarize: if I were a hosting provider or a multi-branch SMB and were to buy a new datacentre strategy right now, I 'd definitly would have a look at CloudSystem Matrix. If it is not that urgent, I might wait another year for SystemCenter 2012 stack and find a convergence between these two. But the Cumulonimbus and its "vertically developed" idea are definitly at its place here.

One thing I neglected to mention and really is important is that HP now knows the meaning of the word OPEN. HP CloudSystem Matrix is vendor independent through the use of OpsWare. This results in an opportunity to implement it in legacy environments and be fully supportive to different hypervisors, operating sytems and hardware. It also keeps you in from vendor lockin in a new solution.

Yet another DataBase? EnterpriseDB raises the bar

2011-12-05T01:00:00.001+01:00

Sometimes you meet people so passionate about their product that they really deserve a platform. I had this feeling when I met Sean Doherty (@OpensourceDB) at #HPdiscover in Vienna last week. Sean is Vice President, Business Development for @EnterpriseDB.

EnterpriseDB is a PostgreSQL Database solution and was invited to HPdiscover as a finalist in the HP Mission Critical Innovation Awards in the category Best New Application.

EnterpriseDB just missed the award but because Sean is such a nice guy, the bloggers awarded him with the "Bloggers GreenIT award for 2nd Losers" at the dinner table (picture). I am so sure the smile on his face was genuine! :-)

During the event I had the chance to talk to Sean a little deeper into his passion. Although I am really not the Database guy (almost all of my work involves iron as a base). I did have an angle to ask him some questions.

Q: For starters what is EnterpriseDB?
EnterpriseDB's Postgres Plus Advanced Server is the only Open Source enterprise solution based on PostgreSQL. One could say that EnterpriseDB is for PostgrSQL what RedHat is to Linux.

Q: I know Microsoft SQL and sometimes we see Oracle DB. Why would I want to use EnterpiseDB?
PostgreSQL has already been recognized as the most advanced Open Source database worldwide. We just took it to the Enterprise level but without the costs one has to pay for the legacy vendors.

Q: you just hit a nerve: what about that cost savings?
Postgres Plus Advanced Server is sold under a subscription model, and licensed per socket. We do not 'punish' our customers for taking advantage of hardware advancements or even use of virtualization. Regardless of the number of cores, and no matter how many virtual machines our customers are running, they will never pay more than the number of physical sockets on the machine. This provides a very low cost, predictable model for database costs.

Footnote Hans: this means if one would have 8 virtual EntepriseDB servers on 2 Hypervisors with each 2 sockets one would only purchase 4 licenses. And on the other side if one would have 16 hypervisors with 4 sockets but only 2 EnterpriseDB virtuals, one would only have to buy 2 licenses.

Q: Ok, I have the "getting it sold to the customer" angle. How about "getting it sold to the engineers"?
That's a good question. Besides the fact that customers want low cost solutions they also will not sacrifice on performance. And certainly not on "ease-of-use". In fact it is at this point EntepriseDB took it to the next level. We added 4 different area's to the community version of PostgreSQL:

TOOLS: on of the most recent launches was the Enteprise Manager, which is the first truly commercial grade management, monitoring and development tool build for Postgres databases
SECURITY: PostgreSQL is already considered to be the most secure database in the world, but we add more for our enterprise customers to ensure the absolute highest levels of security which lead us to EAL2 of Common Criteria Certification
PERFORMANCE: lots of features such as DynaTune (to dynamically tune the database server) and Infinite Cache (to store data in memory for staggering performance gains on read-only or read-mostly workloads) are just some examples that are added in the Advanced Server.
ORACLE COMPATIBILITY: With our 7th generation technology, our customers can migrate an application written to Oracle database over to Postgres Plus Advanced Server with minimal code change to the application. This is very important for the HP Itanium customers now that Oracle went on a rant.

Q: does this mean you are focused at Oracle and Linux?

No, not at all. We support Linux, HP-UX and Windows (32 and 64 bit). We also support all major Hypervisors such as vSphere, Hyper-V, XENserver, KVM, RedHat, ...

To wrap it all up: I think the EnterpriseDB Odyssee has just started!. I will definitely have a look at it. Good luck to you and we will meet again.

LINKS:

EnterpiseDB: Website | Twitter

HP Open Source Database Solution Brief

I also found this video of Jake Ludington (who also joined us at HPdiscover) interviewing Sean earlier this year.

HP discovers the sound of music

2011-12-04T20:22:00.001+01:00

A few times a year I try to join some tech conferences from vendors we work with. So that's especially DELL, HP, Microsoft and VMware. The only downside on being an EMEA citizen is that we get the smaller version of those conferences half a year later. Nevertheless #HPdiscover Vienna was on my schedule from the moment I heard about it. For the first time HP joins all its conferences into one big event where they show everything you can imagine.

At a certain point in time I had to reclaim my request to join the conference at my managers office. I got invited by the HP Social Media team as an "independent blogger". I must admit that the first time such an invitation comes in, you go home pretty proud about it. In fact, I finally got a decent explanation for my wife why I sit behind my laptop, iPAD or just smartphone for hours after she goes to sleep :-)
Just one year ago, I was at an HP Technical Conference in Istanbul after which I ranted a bit about their lack of Social Media awareness at this blog post. I certainly hoped this time to see something different.

So what was different being at Discover as a blogger instead of a reseller. EVERYTHING. Lets start of with the travel, hotel, transportation, ... OK, lets not. We're not here to brag. But HP did a hell of a job keeping the "independent" happy. The nicest thing was our winter jacket goodie (picture above). But if you know which bloggers were at the venue, you know keeping them happy will not buy them out. Bloggers like @nigelpoulton (Infosmack Podcasts& TechnicalDeepDive), @knieriemen (NekkidTech Podcasts), @chris_mellor (The Register), etc ... are just not that easy to bribe. I even don't know if all of them were there at HP's request or at their own initiative.

Nevermind all above mentioned things. Here are some moments of a blogger at HP Discover. Definitly worth doing over! You want to know who else was there with us? Most of them are mentioned in this blog by Calvin Zito (@HPstorageGuy).

The bloggers lounge (aka Bloggers Penthouse). In the front you see @GeeksRoom our "Mexican Racist", who is for starters not Mexican at all :-)

Besides an awesome espresso machine we also had an Espresso bar with a "Barista" (Italian for Bartender)

The serious stuff happened over here: our very own coffee room for private (sometimes NDA) sessions with the HP speakers and loads of time for Q&A. Every morning on our arrival at the venue we received a schedule for the day with the most interesting sessions already picked out for us.

Here 2 HP employees at our Coffe Room Chats: Rafal Los (@Wh1t3Rabbit) with a session on Enterprise Security and Paul Muller (@facne) talking about Software Development (DevOps) at HP.

Dinner with the bloggers in one of the oldest restaurants in town. It is here that we learned some really interesting differences in culture. One example was that the US guys had noticed I had a beer during lunch and told me that would result in losing my job if I were in the US. Out of curiosity I asked what they drink at lunch. The answer was "soda". To summarize: they drown their food in sugar, I use alcohol. Now make your own choice ;-)

The days were long, we did work a decent schedule so sometimes some people get that tired that they fall asleep on their fork and have their hair lying in the whipcream.

The next day some of us got smart and stepped by the VMware relaxation booth. From the left to the right we see Michael Letschin (@MLetschin), Greg Knieriemen (@Knieriemen) and Chris Wahl (@ChrisWahl) taking five in a massage chair.

And last but not least: Calvin Zito playing "the Piano Man"

Active/Active multi-site NAS with HP x9000

2011-11-11T09:31:00.001+01:00

Sometimes you got to answer specific questions with solutions that do not exist (in your portfolio). This is one I got this week:

Prerequisites:
* streched NAS solution (homefolders for VDI)
* active/active on failover

Background information:
* 2 sites close to each other with dark fiber in between

The fist challenge is finding Active/Active stretched NAS clusters. They just do not exist. There are single namespace solutions that will reopen your session but the files will be closed and reopened. So this is not an option.

Now think out-of-the-box ... there is a dark fiber and the customer tells you there is 'no' latency between the sites. This opens doors! Why don't we see it as one dataroom with 2 racks? This makes it easy: take any HP x9000 cluster and your set. If one controller fails, no sweat. But what happens if half of the enclosures fail? Ouch again. So now we need to find a solution to lose half of your disk enclosures. Mmm ... wait a minute, we got P4000!

So here's the solution we came up with:

DISCLAIMER: this is not (yet) approved by HP technicians so is not really supported at this time. We still need to look at the maximum latency these 2 solutions can handle. Will follow up soon.

UPDATE: the design has been disapproved by an HP technician at HPdiscover. It appears that the x9300 Active/Active gateway relies on the iLO to stay online which would not result in a failover on a site-to-site link failure. My point of critique: what will happen on a MoBo crash? iLO dies too in that occasion.

DELL Solutions Conference (channel summit)

2011-10-23T22:31:00.001+02:00

A big part of my new job as Solutions Team Lead is 'Vendor Management'. Therefore I could not mis the invite from DELL to join them at the EMEA Summit for their Preferred and Premier Partners (@ Barcelona). As a fact I can proudly announce that FERRANTI (my employer if you didn't knew yet) is the first channel partner for Belgium that received the status of Premier Partner.

Back to the summit ... What is the main reason Dell invites us here? Well the intention is to 'listen'. It is something you hear a lot from DELL and they do a great job at it. Here are some of my notes of what we heared.

===================================

DAY 1

-----------------------------------------------------------

Introduction keynote:

A few years ago DELL was a direct selling company. So what happened since they started the channel program. As it turns out 28% of commercial revenue already comes through the channel partners. Thats almost a third! Another number is growth: DELL has doubled their revenue (in EMEA) in the last 4 years and half of that growth came through the channel partners. What to expect? Double this total again in 2 years and x5 in 5 years.

So how are they planning to do this with the channel? Dell wants to focus on the co-selling between direct and channel program (nice to hear that being confirmed) and will provide us with multiple new tools to get that accomplished. There is much more than this but not really for a public blog ;-)

Q&A:

Q: what is the strategy for direct sales now that DELL shifts a part of the focus to channel sales?

A: we should not look at direct sales as a threat. On one hand DELL just does not have all in place from a services point of view where you as channel provide much more benefit for the customer. On the other hand it is an advantage for the channel partner to work do co-selling. (important side-note: commission for direct sales does not change if the deal goes to a channel partner!)

Q: if channel already has a 28% share on revenue, whats the limit here for DELL?

A: We are on our way to a 50/50 balance in the next few years but from DELL's point of view there is no ceiling.

Q: where does DELL see the social media as a marketing tool?

A: actually we just started using it just to get more information faster and directly to end-users than going through the traditional marketing stack.

Q: why isn't DELL positioning the channel program more to the market (for example on front page of website)?

A: DELL has done some efforts lately on doing big events together with the channel such as Storage Forum or DELLworld and future enhancements are there to come.

-----------------------------------------------------------

Growing your business with DELL:

If we look at IT budgets we notice that customers spend to much money on IT and of that money to much is lost in 'keeping the lights on'. We should change our way of thinking from; "give us your mess, we'll deal with it" to;

- get rid of avoidable mess

- what you cannot avoid > automate

- what you cannot automate > get it in the cloud against a decent SLA

Q: (from Poland) is DELL planning to go direct in countries it is not active now?

A: NO. We have a decent channel program in place that works fine.

Q: DELL is growing with its own datacentre solutions? What will this mean in services?

A: We will need people to handle the datacentre services, but "people" will never be a product.

Q: what is your point of view on the HP PSG story?

A: as there is no official story yet we can only confirm that DELL will continue to invest in the FULL range of technologies.

Q: what will DELL do on services for new acquired technologies?

A: actually the same answer as on the first question. We can state that: if we do something direct where we shouldn't have, it's a screw-up, not a conspiracy. (#lol)

-----------------------------------------------------------

Impact of trends in IT - by Clive Longbottom (independent IT analyst) - @clivel_98

Im not going through the entire session of him as Clive is not really a DELL speaker. The topic of this talk was ... drumroll ... the cloud (again, yes). There are a few key point I'll shortly address:

A big question of integrators is if the public cloud will be the only channel in a few years. No it will not. There will always be a need for local integrators to handle migrations and links between private, shared and public clouds (the hybrid cloud). So what is going to change? The way we purchase and sell everything. Sales commissioning for example is not going to be one time deals anymore with some licenses and maintenance but that will evolve in long term partnerships and more deals that are spread over 3 or more years.

At the end of his session there was also some time for Q&A. Please let me express upfront that these are his personal opinions as an 'independent' speaker!

Q: as an independent analyst, what can you say about HP as a company?

A: when Mark Hurd came along he actually did a better job then I expected him to do. But Leo Apotheker was a software guy in a systems environment and that just didn't work. In fact if you look at the last 7 CEO's HP had there was not one transition that held some strategy behind it. Losing Apotheker is not enough, they should have let go the entire board...

Q: so what do you think about DELL then?

A: if you would have asked me this a few years ago I might have answered the same thing. DELL was a cheap product selling company that was disruptive and undercut the market and I didn't like that. With a changing market DELL has grown out of that spot and focusses on solutions rather than products especially with all the acquisitions of the last few years. I am impressed how DELL plays its professional services.

===================================

The closing speaker was a pretty technical (product) session of Intel. I do understand that sponsors deserve a spot on an event but this session was not really what i was expecting today.

===================================

DAY 2

-----------------------------------------------------------

After a good night sleep (ahum ahum) I was one of the 25% that came out of their room the fist few hours :-) However they did not really miss a lot those first sessions. DELL put 5 'canons' on stage, each in their territory with each definitely have something to say. The only problem is that the way they told it was like a "political pre-organised panel show". All questions and answers were prepared. I could have gotten the same information through an email, websites or twitter. Ow right ... I allready did.

-----------------------------------------------------------

We welcome a well deserved coffee break (coffee on the table in the first sessions would have been a blessing) and go into separate sessions per country. The meeting is led by our Belgian Channel director. After getting used to being with your biggest competitors on the same side of the table, looking at how to do better together we had some very good conversations. These are some of the ideas (I wont give them all away off course :-)

* we should do more vendor/channel marketing throughout the year

* we should be ready together whenever new solutions/products are launched

* we shouldn't sell 'iron' but solutions

* we need more 'vertical' solutions from DELL (products within a solutions, ex. vStart)

===================================

SUMMARY:

DELL comes a long way as a direct selling company for us who are used to work with vendors but they do a great effort. Especially for those people who have never seen channel there is no better way to end my blog than with a quote I got from one of my 'competitors'

CHANNEL IS THE CHEAPEST SALESFORCE FOR ANY VENDOR

DELL Equallogic Models 2011

2011-08-25T20:55:00.005+02:00

Last week DELL announced a new generation of Equallogic models. I think everybody knows the Equallogic concept by now: if you need to expand, just buy another member, put it in your rack and add it to your storage group in 5 minutes. And then? ... Nothing. It just relevelles itself, distributes all volumes over all group members and if you have multiple disk or group types it will even do sub-lun-tiering for you according to the load of the data.

But what about those models. What fits best for me? Well, you got the models that are different by connection (2 x 1GbE, 4 x 1GbE, 2 x 10GbE), amount of disk slots (16, 24, 48) or drive type (NL SAS, SSD).

So I made an easy overview with the mostly used models (I left the 300GB versions out for example).
You'll see that I did not mention the RAW capacity but the netto capacity according to best practise (RAID50 with 2 hotspares).

Edit: had a conversation with someone at Equalogic and he asked me to clarify that the XS/XVS is not 2 volumes as it might seem in my designs. It is one big volume. It is also not a RAID50 as all otherS but a RAID 6 with one hotspare.

One thing you should have noticed from the picture: the PS6100 controller finally got an OOB (Out-of-band) management port. This way you don't need to give up one of the four iSCSI NICs. And if you did not want to give up that 4th iSCSI port (always in our case) you needed a management server in the iSCSI network. So Thanx for listening DELL (again).

Another positive note: all new models are already in VisioCafe!!!

Vendor Acquisitions & Partnerships - V2

2011-07-30T15:09:00.002+02:00

Last year I blogged Vendor Acquisitions & Partnerships. The main purpose of that blog was to give myself an overview where all that technology we use on a daily base comes from. Almost a year later, it's time for an update. Maybe this time I can give some of my own insights on what's happening out there.

So here's how the world looks like today:
UPDATE - 8/9/2011 - HDS Buys BlueArc
UPDATE - 26/11/2011 - LSI buys SandForce (SSD)

What happened?

HP did no big acquisitions. But they did loads of other things that are worth mentioning. For example: StoreOnce ! For the first time in years HP announces a product that comes out of their own labs. And they nailed it. StoreOnce gives you the ability to dedupe blocks up to 20:1 resulting in only 5% of the original storage size. Now that gives us finally the opportunity to keep our backups on disk for a longer period of time (or even forever!). Is that all they did? No, HP is strongly working on integrating everything together (the Converged Infrastructure story) and using their knowledge in world leading legacy hardware to use that to create new solutions. This results in solutions as the e5000 and x5000. Two differerent solutions on identically the same hardware: a 3U box with 2 blades and a 16-disk cabinet. The first one ships as a pre-installed Exchange Cluster, the second as a pre-installed Windows Storage Server cluster. Nice job of the solutions teams.

DELL is on a roll. In 2007 Michael Dell said he was going to the Enterprise market. Forget about the "online-PC-shop only" strategy, we're going big! The acquisition of Equallogic was the first step in that direction. After losing the 3PAR-war last year to HP ($2.4 billion), they didn't let that stand it their way. Before 2010 was over, DELL bought Compellent ($960 million). If you didn't knew better, you'd say they fought that war with that in mind ... What's next on the list? Well, we don't have decent networking yet. The PowerConnect is a good product (especially together with the EQL) but its not the data-center product. So everybody expects DELL to buy Brocade, their OEM partner for a long time now (the other one is Juniper but we know where that resides now). And what happens, they buy Force10! We'll see what that will give in the nearest future but this really makes that step into the Enterprise Class reality. So what does DELL do with those acquisitions? Well, last year DELL bought Ocarina (software dedupe) and Exanet (NAS). This already translated into the new FS7500 NAS head for the Equallogic family. 2 X 1U High available NAS controller with a 2U battery. It's based on the Exanet software and I've been told that for the next update release the Ocarina dedupe should be a new feature of this same appliance. Besides acquisitions DELL still has some OEM partnerships such as Juniper appliances, IBM tape library and the EMC DataDomain and NSxxx FC arrays. However DELL finally came up with a version of de MD3600 with FC controllers (MD3620fc). Together with the Compellent arrays this could lead to an end of the DELL EMC times.

EMC is the 3rd big boy out there. What do I love about EMC? It's a solid company making solid storage solutions for decades now. But what they do with their acquisitions is not really my kind of thing. And the communication around it is even worse. Let me explain: EMC announced the biggest event of the year and that they would be breaking records! Did they lie? No, they put more people in a Mini than anyone has ever done before. And some kind of Evil Knievel Jr. jumps over more Storage Arrays than anyone has ever done before. Is that what we were waiting for? No, give us new products and solutions. Ok, we'll give you over 40 new products! Waaw you'd say? Nope, again disapointment. The new products are jus the old products with a new front cover (VMAX). Nothing more nor less. The only new product I saw was the VNXe. And then they showed DataDomain and Isilon. Whats new? Nothing. It's the same products as they were before. Both great products but what a fuss about nothing? I'll even go one step further (this is the part about handling acquisitions). If you were an EMC partner you are not an Isilon partner and if you were an Isilon partner, you're not an EMC partner. Say what ?!?
A last example was the benchmarking BS. For decades now EMC has rejected all open benchmarks. Suddenly they announced to join the SPC and shortly after that they revealed breaking all former SPC records. What they did not mention was with what solution they did that! Hundreds of SSDs on several arrays ($/IOP? - $/TB?). So the only thing that happens here is that EMC gives itself some free publicity on technology websites that don't think that far. And if you pop up that question the only answer they can give is: that's why we don't rely on open benchmarks.
My answer: if you're not making crap, don't sell BS!

At Hitachi, IBM, Cisco and Oracle there is not so much movement, or not really in my area of work so I can't really make an opinion on these.

Citrix did a new acquisition this month: Cloud.com There was not that much rumor about it on Twitter so it seems not to be that big deal. If you look at what this does it triggers me anyhow. Cloud.com is one of the founders of OpenStack.org on wich DELL (link) and HP (link) have announced this week to be fully supportive.

NetApp (aka NotApp). Ok, I already sold my opinion between the brackets :-) I'm really sorry but I cannot take this company serious. First they brag for years that their one-size-fits-all is the best solution in the market. Why would you do storage tiering en virtualization on block based storage if you can have NetApp? And guess what? This year they bought engenio (an LSI company). Euh, say what? Yes, a SAS, iSCSI and FC frontend array. By the way the only solution i know in the market that does SAS direct attached & Network based presentation together at the frontend. One should think they might lose the arrogance a bit. Nothing of that kind happens. Every few weeks some NetApp blogger attacks one or all other vendors with loads of crap/fud. I'll just give one example. Early 2011 NetApp announced they were the biggest growing company (we're smashing all competition!) in the storage market with a growth of +35%. Big applause please! Well guys, did you know that HP as a company grew more last year that NetApp is even worth? Wanna play with the big boys? Grow up instead of growing out.

What's still out there?
There are still some smaller players available. Therefore I added a block of "free-agents" that can be drafted (did I mention I like Football a lot). Most of them were already in my last years blog but I updated it with some other storage software solutions.
But there is something else besides the storage solutions an-sich nowadays. We already managed to get higher volumes on a small footprint (larger disks, no-zero technologies, dedupe,...). We also managed to get more IOPs out of that same footprint by adding SSDs to the array. But somehow at some point that is not enough. For larger VDI or SAP environments you'd need tons of SSDs to provide the necessary IOPS. There are some companies out there such as FusionIO and XSIGO that will handle those requests before they hit the array. Definitely worth looking into and possible acquisition targets.

What's Next for the big boys?

* HP & DELL: besides pushing your own strategies on management (ex. DELL Kace), make sure you are completely ready for Windows Server 8, Hyper-V 3 and System Center Stack 2012! The first one who can monitor AND manage all its servers, storage and networking hardware here will have a huge advantage over the other. Right now, most integration packs are not making it.

* DELL: give us a decent alternative for block based dedupe & replication (such as HP StoreOnce and EMC Datadomain). After that, cut the bonds with EMC, they've got nothing you can't do on your own.

* EMC: I'll just repeat my last line in your praragraph: If you're not making crap, don't sell BS!

LINKS:
HP StoreOnce
HP Converged Infrastructure
HP E5000 - Microsoft Exchange in a box
HP X5000 (G2) - Micosoft Storage Server 2008R2 cluster in a box

DELL Compellent
DELL - Force10 networks
DELL FS7500 - NAS head for Equallogic
DELL MD36x0f - entry Fibre Channel storage

EMC VNXe
EMC DataDomain
EMC Isilon

Citrix Cloud.com
OpenStack.org

NetApp Engenio

FusionIO
XSIGO

FOOTNOTE: If I say something wrong, please correct me. If you are offended, ask yourself why!

Equallogic Multipath driver for vSphere (MEM) installation

2011-07-16T11:16:00.000+02:00

vSphere is known to work out-of-the-box. Dell Equallogic does this too. But as it is not Microsoft, nobody stops at next-next-finish ;-)

Whats not to forget? Multipathing, multipathing and multipathing. Let's start by showing what a mounted Equallogic volume looks like on a nex-next-finish installation:

You should have noticed that vSphere recognizes it's an Equallogic volume (Array Type) and found 4 active channels to the target (paths). It does not however take full advantage of those available paths. In an environment where your array supports Round-Robin (HP EVA, Lefthand, DELL MD3000,....) you could just set the setting for that volume to vSphere Round-Robin:

Result: 4 active paths to the array . Hurray!

Are you pleased? Well, not quit yet. We still need to do this for each datastore on each host. Therefore you could use following simple PowerCli script to set your default MPIO settings to RoundRobin and your Equallogic MPIO settings also to RoundRobin:
esxcli nmp satp setdefaultpsp –satp VMW_SATP_DEFAULT_AA –psp VMW_PSP_RR
esxcli nmp satp setdefaultpsp –satp VMW_SATP_EQL –psp VMW_PSP_RR
esxcli corestorage claimrule load
esxcli corestorage claimrule run

If you think you have a job well-done, there is more to come! What we stated here is basic RoundRobin for any supported array (FC or iSCSI same trick). The DELL Equallogic team went one step further and developped a Multipathing kernel plugin specifically for vSphere called the MEM (Multipath Extention Module).

There are 3 ways to install the MEM:
1) run the setup.pl script from the VMware Management Assistant (vMA) virtual appliance
2) run the script from vSphere CLI 4.1
3) vSphere Update Manager does it all for you.

As you might know by now ... 'does it all for you' suits fine by me :-) Let us show you how simple this works:

1)Start Update manager and import new patch

2) browse for patch and confirm import

3) result: Patch is imported in Update Manager

4) Create a new baseline (I prefer doing this seperated from other updates)

5) give name, desription and select Baseline Type as being a Host Extention

6) add the patch to this baseline (here is the part you could add other updates too)

7) confirm and the result gives you a new EQUALLOGIC baseline for ESX Hosts

8) Now simply go to the server(s) you want do patch, select the update tab and remediate the baseline. Now don't forget that your server will be put in Maintenance mode so FT, HA en DPM will kick in at this stage! After an automatic reboot your host is updated.

End Result:

ALL your (current en future) volumes get the Equallogic Multipath selection by default! No need for doing it manually and no need for the PowerCLI script we used earlier to change the defaults.

FOOTNOTEs:
1) The new Equallogic firmware 5.1 has been released earlier this week but the MEM module has not changed so you still use MEM1.0.1. However it is not yet supported for vSphere5! So watch out for the next release of MEM to be fully supported on your vSphere5 (will follow shortly I guess).
2) One thing that will change in vSphere5 is that we won't have to bind our iSCSI nics anymore from CLI! (remember the # esxcli swiscsi nic add -n vmk1 -d vmhba37 ... ) Here's a nice video by @esloof on that new feature. If you don't know by now how that esxcli worked before, here's a reminder video

Backup Exec 2010 - install Linux Agent

2011-07-15T08:32:00.000+02:00

I'm not that great at Linux OSses (or even just 'Syntax' for that matter) so each time I have to do something I need to go back to previous notes or cursing myself for hours when I didn't do that same job before. Therefore I'm a freak at taking screenshots, pasting in a Word doc and save to C:\procedures :-)

So here's a HowTo for installing the Linux Client for Symantec BackupExec. If you know the steps it takes you no more then 5 minutes.

Prerequisite: mount the BackupExec Install DVD to the Linux server. Most of the time our server is a VM so for me it works best to put the ISO (temporary) in a shared datastore.

Step 1) Install the agent

Step 2) Start the agent

Thats all! Couldn't go easier ... if I knew what to do the first time ;-)

HP EVA Capacity and Performance Calculator - V2

2011-07-14T20:35:00.002+02:00

It has been a month or two since my first version of the EVA calculator. The greatest benefit is that I don't get a lot of questions anymore :-)
Link to previous post with ALL details explained (may 4th 2011)
Now why would I update it? Well, no matter how good your tools are, there is always room for improvement. This one is a best practise for the storage administrator and not really for the designer. There was one thing I didnt do before and it has to do with the occupancy alarm. The occupancy alarm is calculated according to the amount of disks and failover level. It shows you how much you can allocate in a pure healthy state (so that disks can fail before breaking de vRAID open). What I didnt do is giving the available free space until you hit the occupancy alarm level. So thats added here:

Another small detail is the currently allocated level. I don't really think this one will help you technically but I needed it in my screenshots for a expansion report.

Download EVA Calculator HERE

Footnote: very proud to have sold my first P6300 (new EVA) yesterday so there might come new updates in a few weeks.

#vTax FUD real life example

2011-07-14T08:50:00.002+02:00

It now has been more then 24hrs since the vSphere5 announcement. Allthough VMware has done tons of work on the arhitectural level wich we all are very pleased with, there is something out there that throws a big shadow: the new licensing model, a.k.a #vTax (lets make the name official shall we?)

I'm not going to repeat what everyone else allready said. The real VMguru's will defend the strategie as far as they can, the partners (hardware vendors) and the resellors (as I) just don't get it. Allthough you'll find some blogs that state "it isn't bad at all" (sorry @gabvirtualworld) you just can't defend this strategie all way trough in my opinion.

So why do I feel I have to say something too? Well, this morning I read Calvin Zito's (@HPStorageGuy) blog on their new support for up to 2TB in 1 server! (blogpost). The last few years all hardware vendors have been working their asses off to save footprint, power, cooling, blablabla,... And once they get there you'll have to pay +150k per server just for your VMware Licenses? So now the answer of the beleivers could be: who the hell puts 2TB in one server. Isn't that a little bit to much showing off? Ok, I hear you. Therefore I will give you an opposite example.

Lets say you don't really need an enterprise environment with tons of hypervisors. Could we still take the benefit of the architectural strengths of VMware? Oh yes you can. VMware launched Essentials+. You get vCenter, HA and VMotion, almost for free. For free you say? Yes, only 3.500. The only limitation was max 6 sockets. How do we size this? Well, you take an HP DL380 (or DELL r710) with 2 quad or 6-core sockets and each socket gets 6 x 8 GB RAM. Not really oversized no? Result? 3 x 96 GB RAM per server = 288 GB RAM. Together this means: 12.5/GB on vSphere licenses. And what does this mean in the new licensing model: Essentials+ has a max of 144GB in the vRAM pool so I will need to go to the Enterprise+ linceses becasue of the 48GB RAM per socket. 6 sockets Enterprise+ * 3.500 per socket = 21.000 (73/GB) for the same amount of RAM I allready put in there. Are you kidding me ???

Now lets hope that I will be proven wrong...

To summarize: I think my boss is really happy these days that we have always supported a multi-vendor vision (ex. HP/DELL) as VMware now gave our Hyper-V Consultants job security for the next months.

HP EVA Capacity and Performance Calculator

2011-05-04T17:00:00.003+02:00

NEW BLOG UPDATE + VERSION 2 | JULY 14th 2011

A colleague comes in and asks you what IOPs a specific array can handle. Or how many disks they need to add to supply enough power to handle the current load. Or how many disks they need to add for 5TB extra space in RAID 5.

Having to do this over and over I started to use an Excell sheet with basic calculations. Then pops up a next question: What formula did you use? So I just gave them these answers: Occupancy_Alarm = 100 – ceiling { [ max ( PDM_capacity, HP Continuous Access_log_sum) + 5GB / total_disk-group_raw_capacity ] }

Ok, that might not be the best idea after all :-) A good engineer is a lazy engineer. Therefore I updated my Excell file with as much information as I could get and made it cleaner to use. Now they can calculate everything themselves.

you can downoad the file here

All values are explained and all formulas are public (not hidden) s you can copy paste wherever you need.

Whats in the document:

Disk Group parameters:

DiskType: informational field - not used for calculations

DiskCount: speaks for itself

DiskSize: RAW disk size as mentioned on the disk

(formatted size is in the capacity calculator)

RAW IOPS / disk: calculate on the right or take the averages from the fields comment

Failure Protection: Protection Level set in Command View: single = 1 / double = 2

Read/Write: percentage ratio (can be calculated via EvaPerfMon)

Safe IOPS calculator:

Max Raw IOPS = #disks * RAW IOPS/disk

Safe IOPS = RAW IOPS / {(write% * RAID penalty)+ read%}

RAID PENALTIES:

RAID 6 = 6

RAID 5 = 4

RAID 1 = 2

RAID 0 = RAW IOPS

Capacity Calculator:

RAW Capacity = { #disks * disk Size * 93,13% (formatted size) } - 2 * Protecion Level

Netto/RAW capacity:

RAID 1 = Netto * 2 (mirror)

RAID 5 = Netto + 25% (4+1 parity)

RAID 6 = Netto + 33% (4 + 2 parity)

RAID 0 = Netto

Occupancy Warning Level:

Occupancy Alarm: 100% - ceiling { (max (size of PDM;ConAccessLog) + 5Gb ) / Raw Capacity }

Max Size PDM: 2 * largest disk size * failure protection

Continuous Access History Log: see max size in your configuration

IOPS/Disk calculator

This one is generic (can be used for every physical disk)

Maximum rotational delay is the time it takes to do a full rotation excluding any spin-up time. So 60.000 (amount of ms in a minute) devided by the rotations per minute.

Averge rotational delay: max divided by 2

IOPS per disk: 1 second (1000 ms) divided by the total latency

total latency: seek time + average latency

Disk Counts:

How many disks do I need to buy?

DiskCount= ((UsableV0 * 538) + (UsableV5 * 673) + (UsableV1 * 1076))/(DiskCap * 476) + (ProtLevel * 2)

=================================================================

DISCLAIMER: this is a helpfull tool but not bulletproof! I am still looking for updates/flaws.

=================================================================

Whats next? I'll definitly update this worktool with other arrays I frequently work with such as P2000 / P4000 or even Dell Equallogic. So to be continued ...

ChunkWhat? Chunklets!

2010-12-20T22:37:00.002+01:00

People who know me or follow me at twitter will allready know that lately I have been pretty active with 3PAR. From a HP Partner perspective I'm a bit proud to be an 'early adopter'. One of the (dis)advantages of being an early adopter is that you may explain it all to your friends and colleagues.

Although there is a lot of information on the 3PAR website or at 3PARtv (Youtube), there are still some things that need to be explained 3 times before they (you) get it ;-)
This happens to be the fact for the 'chunklets', logical disks and how 3PAR handles failover on disk/chassis level. Therefore this 'Chunklets for Dummies'.

INTRO - What is a 3PAR array:
A 3PAR array exists of 2/4/6/8 controller nodes, direcly connected through a passive backplane. Eachcontroller has redundant connections to pairs of drive chassis. Each drive chassis holds up to 10 magazines of 4 disks. You always fill a magazine per 4 disks and 1 magazine per chassis (4 chassis = 16 drives minimum).

1) What is a chunklet? In short, a chunklet is a 256Mb block of data. By default a 3PAR array will devide ALL DRIVES in chunklets of 256Mb. This means for example for a 600Gb drive you'll have 2400 chunklets (600Gb*1024Mb/256Mb). If you now take a basic 3PAR array with the minimum of 16 drives you would have 38.400 chunklets in total.
From here on I will use this graphic that gives you a view of a 4 chassis array, each with 1 magazine of 4 disks and I have drawn 30 chunklets per disk.

2) Spare chunklets: as you might have noticed I allready coloured some chunklets at the rear. These are the 'Spare Chunklets' (in de pictures you'll see them under 'SC'). By default the 3PAR array devides its space into regular and spare chunklets. These last ones will be used for example when a drive/magazine/chassis fails, it are also these chunklets that will be used for dynamic optimization, ...

3) Logical Disk: now we can create Logical Disks. A logical disk is just the amount of chunklets we need to fill up the minimal block with the RAID level we chose. For a RAID 5 (3+1) this is 4 chunklets. So a logical disk of 5 (3+1) is always 256 Mb * 4 RAW capacity (768Mb netto). For the people who don't know what that P stands for: that's the Parity bit which is used to calculate the data on the other members of the raid when one of them gets lost. As you might have noticed on my drawing the 3PAR array by default will stripe these chunklets over max. 1 disk per magazine and per chassis. This means that at any point of time you can loose eighter a disk, a magazine or even an entire disk chassis.

4) RAID 10/50/60: actually this title should be "Common Provisioning Groups' or 'CPGs' but that doesn't read that nice. If you look at the picture closely you will see that I filled up the disks with 6 LD's at raid 5 (3+1). Together these LD's are grouped in one CPG as one big RAID 0. So in total I have a RAID 50 of 4.608 Mb (6 x 768Mb per LD) with 24 cunklets. Now take in mind that a Logical Disk in a RAID 1 (mirror) takes double the amount of space RAW then you get Netto (next picture). You will see that we needed 24 chunklets for the 4.6 Mb on RAID 5 (3+1) so we will need 36 chunklets for 18 Logical Disks with each 256Mb netto or 512Mb RAW.

5) Virtual Volumes: the last stage is creating LUNs (3PAR calls these Virtual Volumes) that can be presented to the hosts. What you see in the last picture is a RAID 10 volume of 3Mb (so which takes 24 chunklets RAW). I think you get the picture by now.

Now multiply all these numbers in real live and you have a decent 3PAR array. Once you get all this, you will be able to adapt all other information on 3PAR technologies like dynamic and adaptive optimazation or even the basics around the 3PAR Asic Mesh.

Now lets hope that pointing to this blog will solve a lot of those explanations I mentioned at the start of this post.

HP ETSS | The day after insights

2010-12-06T23:59:00.002+01:00

As some of you might have noticed, I spend my last week in Istanbul (Turkey) for the "HP Enterprise Technology and Solutions Summit". The annual link between HP and its partners. As there were several sessions under NDA you will not find all sessions written out here but I'll try to give a brief summary of my personal comments.

The humble mastodon:
A first personal comment is that I have never heard an HP as humble as last week. Let me give you some examples:
* "We know we have been giving training today for yesterday's technology and we apologize for that." - WAAW!
* Trougout this week almost all speakers had the same vision when it comes to competition: "never get into a head-to-head comparison but try to focus on the strengths of your product. You never know if the competition changes features tomorrow."
* 2 important statements to follow: "there's no one size fits all" - "there is no product that is the sum of all perfect features"

The monolithic HP:
HP is a monolithic company selling modular solutions. It's a bit a bold statement but I'll try to explain; I have been surrounded by 800 people fulltime involved in ICT for 4days and not once I heard TWITTER being mentioned. Not one speaker started his session with; "hello, blabla is my name and you can follow me @blabla".
And if the speakers don't mention twitter, why wouldn't the attendants do? Therefore I started the #ETSS and #HPETSS hash tags to try and find some 'colleagues'. NOT ONE for a whole week!!! And when I tried to add the word twitter to the conversation, I was called the geek...
So isn't HP out there? Yes they are. I follow several HP Twitter accounts but they do not have a person behind it. They are advertising/marketing accounts.
There are exceptions ... @HPStorageGuy (Calvin Zito) and recently @HPsisyphus (f.k.a. @3ParFarley) (Mark Farley) who both do a hell of a job out there in the social world. When I mentioned these two, there were called 'the Mavericks" of the company. I can assure u that is a compliment!

3PAR - the expensive daughter?
When HP and DELL had their acquisition war over 3PAR 2 months ago, the majority thought HP outbidded DELL just to 'show off' they are still the bigger brother. I was convinced there was more about it before this week but still needed some further arguments. And arguments I got! Not only is the 3PAR solution what it promises to be, HP takes it to the next level! HP already invested the same budget in R&D for 3PAR then id could for itself for the last 10 year. Definitely a story to follow up (already working on the first bid!).

My special thanks go to the entire 3PAR team (Fred, Sjoerd, Lee & Christophe) for the perfect technical sessions and the beers ath the bar and not to forget Jim Haberkorn for a nice chat at the Dataprotector reception.

Netmask Cheat Sheet

2010-10-15T12:59:00.001+02:00

Netmask              Netmask (binary)                 CIDR     Notes
255.255.255.255  11111111.11111111.11111111.11111111  /32  Host (single addr)
255.255.255.254  11111111.11111111.11111111.11111110  /31  Unuseable
255.255.255.252  11111111.11111111.11111111.11111100  /30    2  useable
255.255.255.248  11111111.11111111.11111111.11111000  /29    6  useable
255.255.255.240  11111111.11111111.11111111.11110000  /28   14  useable
255.255.255.224  11111111.11111111.11111111.11100000  /27   30  useable
255.255.255.192  11111111.11111111.11111111.11000000  /26   62  useable
255.255.255.128  11111111.11111111.11111111.10000000  /25  126  useable
255.255.255.0    11111111.11111111.11111111.00000000  /24 "Class C" 254 useable
255.255.254.0    11111111.11111111.11111110.00000000  /23    2  Class C's
255.255.252.0    11111111.11111111.11111100.00000000  /22    4  Class C's
255.255.248.0    11111111.11111111.11111000.00000000  /21    8  Class C's
255.255.240.0    11111111.11111111.11110000.00000000  /20   16  Class C's
255.255.224.0    11111111.11111111.11100000.00000000  /19   32  Class C's
255.255.192.0    11111111.11111111.11000000.00000000  /18   64  Class C's
255.255.128.0    11111111.11111111.10000000.00000000  /17  128  Class C's
255.255.0.0      11111111.11111111.00000000.00000000  /16  "Class B"
    
255.254.0.0      11111111.11111110.00000000.00000000  /15    2  Class B's
255.252.0.0      11111111.11111100.00000000.00000000  /14    4  Class B's
255.248.0.0      11111111.11111000.00000000.00000000  /13    8  Class B's
255.240.0.0      11111111.11110000.00000000.00000000  /12   16  Class B's
255.224.0.0      11111111.11100000.00000000.00000000  /11   32  Class B's
255.192.0.0      11111111.11000000.00000000.00000000  /10   64  Class B's
255.128.0.0      11111111.10000000.00000000.00000000  /9   128  Class B's
255.0.0.0        11111111.00000000.00000000.00000000  /8   "Class A"
254.0.0.0        11111110.00000000.00000000.00000000  /7
252.0.0.0        11111100.00000000.00000000.00000000  /6
248.0.0.0        11111000.00000000.00000000.00000000  /5
240.0.0.0        11110000.00000000.00000000.00000000  /4   "Class E" (Multicast and reserved)
224.0.0.0        11100000.00000000.00000000.00000000  /3
192.0.0.0        11000000.00000000.00000000.00000000  /2
128.0.0.0        10000000.00000000.00000000.00000000  /1
0.0.0.0          00000000.00000000.00000000.00000000  /0   IP space

Vendor Acquisitions and Partnerships

2010-10-08T09:38:00.002+02:00

Sometimes you just lose it these days. Who's who and why or when. So I made, actually for myself, a diagram of vendors that come across my path. For starters, I hope it is at least a correct diagram and second I would like to keep it up to date when new info arrives. So feel free If you miss some important alliances to fill them up.

Who's Next?
Offcourse It immediatly raises some questions. One of the most important questions (in this diagram) now is who is going to buy CITRIX? Is it going to be Oracle as it was written in the stars or will the Unified Computing System alliance with Cisco lead to an aquisition?

What's next?
As all the hardware alliances now have Computing, Storage, Network and Virtualization technology, what will be next? IMHO It will be archiving. Smart archiving will be the next hot thing in the datacenter. Especially now that all tier 1 storage vendors are using smart tiering for fast/slow/external/cheap storage.
So what are we looking at then? Symantec & Commvault ?

UPDATE 15/11/2010: EMC acquires ISILON
UPDATE 15/12/2010: DELL acquires Compellent (final agreement)

remote install of DPM 2010 agents

2010-08-12T08:40:00.003+02:00

CASE: sometimes the deploy of DPM 2010 agents just failes. Most of the time it is just because the default firewall (of W2K8) doesn't allow you to.do it.

SOLUTION: you can do the installation manually but that implies you need to log on to every server. However; a lazy engineer is a good engineer so we will try to do this remote (from the DPMserver).

PREPARATION:

1) on the DPM server, share your agents folder as \\DPMserver\DPMAgents (C:\Program Files\Microsoft DPM\DPM\ProtectionAgents\RA\3.0.7696.0).

2) install Sysinternals Suite (LINK) to C:\Windows\System32. Add the path to your system variables; go to Properties option in MyComputer, Advanced tab, Environment Variables, Path. Put a semicolon after the last character and add your path to SysInternals folder. Test this by running BGINFO from the command line.

EXECUTION:

Run CMD as Administrator

run following commands:

psexec \\CLIENT01 -s cmd /C net use X: \\DPMserver\DPMAgents

psexec \\CLIENT01 -s cmd /C X:\amd64\DPMAgentInstaller_x64 /q

Now that the agent is installed, you need to tell it where the DPMserver is because this enables some ports on the Windows Firewall.

Psexec \\CLIENT01 -s cmd /C "C:\Program Files\Microsoft Data Protection Manager\DPM\bin\SetDpmServer.exe" -dpmServerName DPMserver

Once this is done the Agent needs to be imported in the DPMserver:

C:\Windows\system32\windowspowershell\v1.0\powershell.exe -PSConsoleFile "C:\Program Files\Microsoft DPM\DPM\bin\dpmshell.psc1" -noexit -command ".'C:\Program Files\Microsoft DPM\DPM\bin\Attach-ProductionServer.ps1' DPMserver.domain.local CLIENT01 'user' 'password' 'domain'"

Your DPM agents is well installed but there is just one other thing we can add to this and that's the enabling of the Windows Backup Server feature on the client (otherwise you will not be able to backup System State and Bare Metal)

psexec \\CLIENT01 -s cmd /C dism /online /enable-feature /featurename:WindowsServerBackup

NOTE: these few lines of script are now a base which I will be using to create a decent script that allows you to do it in one time and for multiple servers. If you just put these lines after each other in a CMD file, it does not go well sometimes.

Disk cleanup via Scheduled Tasks for Server 2008

2009-10-28T14:40:00.004+01:00

CASE: for basic maintenance such as disk defragmentation and disk cleanup we want to use Scheduled Tasks.

Enable Disk Cleanup (not there by default):

1) Disk Cleanup for Server 2008: go to "Server Manager" - "Features" - "Add Features"

2) Select "Desktop Experience"

3) Reboot Server

Create settings for disk cleanup:

1) go to start CMD (run as administrator)

2) create a registry key with your settings: C:> cleanmgr.exe /sageset:n (n - select a number)

3) check all options

Create Scheduled task:

1) go to "Task Scheduler"

2) go to menu "action" and select "create basic task"

3) set name (for ex. "Monthly Disk Cleanup") - NEXT

4) trigger: Monthly - NEXT

5) monthly: select all months / on: last + sunday / set hour (01:00 am) - NEXT

6) "start a program" - NEXT

7) browse for cleanmgr (ex. C:\Windows\System32\cleanmgr.exe)

8) add argument: "/sagerun:n"

REMARK: you could add a task for defragmentation a few hours after the disk cleanup.

set ACTIVE USER to AutoAccept Meeting Requests when offline

2009-09-28T09:44:00.004+02:00

CASE: we needed a resource that could be reserved for LiveMeetings (Microsoft Roundtable). This user also needed to be ACTIVE because the system will be provided with a mini-pc that is configured to automaticaly boot with that user and log on to Outlook Calendar.

ISSUE: An active user cannot AutoAccept Meeting Requests when Offline. this would have the consequence that only when user is active, it will reply all pending meeting requests.

BEWARE: UNSUPPORTED METHOD !!!

1) Change mailbox to an equipment type:

set-mailbox -Identity "Mailbox Name" -Type Equipment

2) Change Access Rights on that mailbox for a different user:

Add-MailboxPermission -Identity "Mailbox Name"

-user "User name to grant rights"

-AccessRights FullAccess

3) LogOn to OWA (with that user)

click on your username (top right) and select "Open Other Mailbox"

Type "Mailbox Name" and open

GO to "Options" (top right)

In the main page at the bottom you will find "Resource Settings"

Activate the "Automaticaly Accept"

Disable Reminders

Disable "Allow Conflicts"

Allow "Recurring Meetings"

Aht the permissions select 3 times "Everybody"

Save and Close

4) Go to ADUC (Active Directory Users and Computers)

Enable the resource mailbox USER

REMARK: especially this last method seems not to be 'Catholic' to me because it is against the nature of a resource mailbox (in E2K7) to be an active user. But Somehow the Exchange Server remembers the resource settings when activated again. Need to be monitored after implementation.

How to deploy the settings of your Wireless Network

2009-08-26T09:44:00.005+02:00

CASE: when you have a Wireless Network in your environment you will have to deploy the settings of it to all notebooks. One of the most common known ways to do this is through Group Policy Objects (GPO) - (TECHNET: http://tinyurl.com/njqoc3 ). But there are some reasons why this is not the easiest way. One of them is that you can not provide your deploy with a static pre-shared key (WPAPSK). So we had to find another way.

Solution: as for many solutions we found it in a command line tool: "NETSH WLAN" . This tool is actually the CLI after the "Set Up a Network or Connection" GUI of Vista but with some more tricks (as always in CLI). For the complete syntax and help of netsh wlan: http://tinyurl.com/o8hc5m

How does this work:

I will not go to deep in all the possibilities of netsh wlan but will just go to how we created our solution:

SYNTAX: netsh wlan add profile filename="file.xml"

This is all you needed to do. And our XML files looks likes this: http://tinyurl.com/log2cw

TIPS:

1) put the netsh wlan command in a batch file and you can run it whenever you want on whatever machine (Vista ans W2K8 or newer). We also implemented this batch file on our Altiris Deployment Server wich made it easier for us to do this for multiple machines remotely.

2) If for any security reasons you do not want to put your WPAPSK key hardcoded in your XML file you could enable all your settings on one machine and make the connection work and afterwards EXPORT the settings to a new XML file (see syntax doc). In this file you will see that the switch will be set to TRUE and the will be an encrypted key.

For all other possible items in the XML file go to: http://tinyurl.com/l8mysr

Heritance of Folder Permissions in Outlook

2009-07-17T16:50:00.002+02:00

CASE: there is a new secretary and she(he) needs access to all inbox folders of her manager whitout having full control on the mailbox object.

ISSUE: everybody knows how to delegate permissions on an inbox or calendar. But did you also notice that it only applies to the folder you selected and NOT for all existing subfolders ? Classic examples: customers / suppliers / helpdesk / accountancy / ... etc. with each another thousand subfolders :-)

SOLUTION: When you don't have the other thousand subfolders you could go over every folder manually. When you don't want this, this is the only way (i found) to get the job done:

1) create new folder [TEMP]
2) MOVE all existing folders under your [inbox] to the [TEMP]
3) apply the permissions to the [INBOX] you wanted
4) COPY all those folders back to [INBOX]
5) DELETE [TEMP]

So what happened? By default the permissions you gave did not apply to EXISTING subfolders but will aplly on all NEW folders.

Easy way to Reboot W2K3 server automaticly

2009-06-16T14:12:00.002+02:00

Sometimes It can come in handy to reboot a server on a scheduled base. This can prevent memory issues (SQL/Exchange/...) and sometimes just to force all services to reboot.

You don't have to be an expert to do this:

Go to "Scheduled Tasks" and create new scheduled task

Browse to folder WINDOWS/System32 and select "shutdown.exe"

select schedule (for example weekly)

and define schedule on next page (ex. saturday 23:30, every 2 weeks)

Confirm the user. BEWARE that this user must have password settings that don't expire!

now on the last page select "open advanced properties". If you don't do this, the system will just shut down.

Add the following to the run script: C:\WINDOWS\System32\shutdown.exe /r /f /t 0 /d p:0:0

explanation of the switches:

/r = reboot

/f = force applications to shutdown

/t 0 = zero second waiting time (you can add time for the user to exit script)

/d p:0:0 = the reason why you shut down (in this case: planned no reason)

Apply again with the same username and password.

For more information about the switches, just run shutdown.exe /? in a command box.