If not, then welcome to the full-blown ScienceLogic water balloon fight, courtesy of the instigator Jeremy Sherwood. In minutes, he transformed from our successful SciLo Cloud Strategist into a merciless water balloon launcher (picture Buddy’s snowball fighting technique in Elf). Out of the blue, among the throngs of people chatting around the patio, Jeremy’s lofty tosses were going in all directions. Nobody was safe: not the innocent corn hole players, or the group comfortably lounging in the shade, or even Josh Taylor, our Support Engineer, with his professional camera. It didn’t take long for retaliation, and war ensued; more employees returned to the office with wet clothes than dry. The highlight? Cody Mruk (inside sales)’s bold toss directly into the center of a ‘safe zone,’ where those hoping to stay dry had clumped around the food tables. I guess the clichés regarding what it takes to be a sales guy are true; “you gotta risk it to get the biscuit.”

Here at ScienceLogic though, we expect everything to be multitasking (this could be a result of our software’s incredible ability to do everything at once). Therefore, our picnic was not just a picnic, it was also a water balloon fight. And our water balloon fight was not just a water balloon fight, it was also a fundraiser. Dave Link, although he mysteriously ducked out of his duty as the sole target of the balloon throwing, still showed his support for the tornado victims in Moore, Oklahoma with a  pledge to match all money raised by ScienceLogic. The proceeds, which added up to a grand total of $1,826, will be donated to the American Red Cross Disaster Relief for the Oklahoma Tornado Relief Fund.

Everyone knows that we will confidently put ScienceLogic network management software up against any rival product on the market, but I am here to inform you that ScienceLogic has a hidden talent: we could confidently field a water balloon fighting dream team.

A failed attempt to catch a water balloon leaves Martin Cooley the soaked victim of a wall of water.

SciLo team members play a round of cornhole

Winding up, Dani Church launches a water balloon back at Jeremy Sherwood

Photos by Josh Taylor

There is a critical incident reported by the ScienceLogic network management system that is being handled by your PagerDuty service.  First, you acknowledge the incident so your boss doesn’t get the next call out.  Next, you click on the embedded link in the PagerDuty incident and it prompts you to login to your ScienceLogic portal on your tablet.  Up comes the source event and you can see the issue causing a potential service disruption.  Within a few clicks, you’re off troubleshooting and addressing the problem while the others at the party wonder where you are.  After a few minutes resolving the issue, a new event appears in your ScienceLogic Smart IT software indicating service is back to normal.  You double check with PagerDuty and you see that the original incident has already been resolved.  It’s back to the party and back to having a fun Friday evening.

The above real-world scenario is now possible with a powerful integration between ScienceLogic and PagerDuty. PagerDuty is a SaaS based alarm aggregation and dispatching service for system administrators and support teams that works anywhere at any time. Combined with ScienceLogic’s robust IT operations monitoring platform for data center and cloud environments, you have a match designed for the most modern IT shops. The advanced integration leverages the robust always-on PagerDuty API with the Run Book Automation capabilities of the ScienceLogic solution. ScienceLogic network management events are synced bi-directionally with PagerDuty incidents with the ability to pass the user name who acknowledged the incident back to ScienceLogic. By offering this seamless integration, teams can trust that high priority events are being tracked and handled appropriately.

The integration does not stop there.  By leveraging our Dynamic App collection of API data, ScienceLogic also provides a PagerDuty service dashboard that can track the number of triggered, acknowledged and resolved incidents as a performance metric over time.  You can also trend the transaction time of PagerDuty requests and the percentage of acknowledged incidents.  For highly distributed IT service desk teams, this dashboard can be used to monitor the effectiveness of your team’s response to major faults and alarms.  With our new partnership and integration with PagerDuty, our customers benefit from knowing their mission critical IT service incidents are being managed by the right person at the right time.

Read PagerDuty’s blog post about our partnership and integration.

The question now is – what is next for both Dell and the cloud computing sector? 

In the latter case, there are increasing rumblings around the barriers to entry being so high, that we’ve seen the last of the great entrants to this cloud sector.   It’s possible to make that argument at the infrastructure layer, but far too premature to make the same gesture at the SaaS, apps, and mobile sector layer where innovation is always seconds away from market adoption.  In the meantime, the company has not given up on its established embeddedness at the infrastructure layer – an area that still has plenty of room for market share acquisition given the perception of its competitors right now.

HP CEO, Meg Whitman, is nonchalant about Michael Dell’s taunting of HP’s losses to Dell in the x86 Server Space.  How can she afford to shrug off the losses?  Staying away from unprofitable deals.  And therein lies the lesson for Dell too.  Dell is currently taking on debt by slashing server prices.  All this while investors jostle around the buy-back of the company – a return from the public sector to being a private entity: a position that offers the luxury of being able to admit mistakes and reduce bad investments without public scrutiny or overt punishment.

This is Dell’s opportunity to dedicate itself to the cloud competitors as a Service Provider to them, instead of competing with them.  In quickly announcing partnerships with the likes of Citrix VDI solution, the company wants to own the next gen IT service delivery space by promoting the revenue generating activities that are required in the creation of the latest cloud platforms, rather than in abstracting the costs and operations of this infrastructure from its customers.  Its latest earnings show that the market is responding by giving it growth in the IT services area.  It’s something that HP also figured out when it bought EDS.  In fact, Dell Services (combined with the old Perot systems acquisition) generates 75% of its revenue from infrastructure and related support services, 18% from applications and consulting services, and 7% from business process services.

Where does this leave OpenStack?

The announcement to detract from its cloud play is further reaching than first meets the eye in that the company is also no longer pursuing its building of a public cloud based on OpenStack (although it will support the stack for private cloud builds).  In fact, the company’s desire to sell clouds through partners was exemplified by initial cloud platform partners Joyent, ScaleMatrix, and ZeroLag – none of whom are OpenStack users. The response from OpenStack has been largely dismissive of the announcement as having a negligible effect since Dell will continue to contribute to the OpenStack community – a material player in the longevity of this cloud OS project.

In summation, this move will actually free Dell up to be more flexible in trying to promote competitors to Facebook, Google, and Amazon, none of whom use commodity hardware for their cloud platforms.  In addition, it means Dell won’t have to go head to head with VMware, Microsoft, and Google, that themselves are brewing toward a cloud collision that will soon be seeing new and improved tools, price reductions, and customer churn.  Dell might well look back a year from now and wipe the sweat from its brow at having missed that close shave.

Example 1:

Hacker notifications via email:  The email is automatically triggered and includes the information below along with a link to the event.

Severity: MAJOR
First Occurred: 2013-05-14 19:27:28 PDT
Last Occurred: 2013-05-14 19:27:27 PDT
Occurrences: 4
Source: Syslog
Organization: DEN Colo
Device: den-srv01-sl

Message: [multiple] Invalid user oracle from 216.69.164.92

Sent by Automation Action: SILO Only

Example 2:

Proactive notifications with any sort of WAN routing issues that would impact our customers (in this case, show attendees).  All of these messages get routed immediately to our Routing Engineering specialist in the NOC:

WAN routing dashboard | Click to view larger image

Example 3:

A new integration with F5:  Dave already discussed this integration on his Interop blog post, but it was an incredibly robust integration done completely on-the-fly at HotStage so here it is again.

DCM view of F5 LTM Logical Structure | Click to view larger image

Example 4:

Node Performance characteristics, such as traffic broken down by IPv4 and IPv6 to Interop.com web site(s):

Node Performance dashboard | Click to view larger image

Example 5:

A live map visual of the status of the Show Network which Will built that was a major crowd pleaser:

Show Network dashboard | Click to view larger image

A big thanks to NOAA who supplies the background for another dashboard from our network management system that was very popular.  This one shows the status of our WAN/Internet connections:

WAN/Internet dashboard | Click to view larger image

Our fifth time managing InteropNet was another success and we look forward to doing it over and over in the years to come!

Here is a roundup of memorable moments from a few of the team members that attended the show:

Rick Larson, one of our newest Account Executives, said, “Interop was amazing!”  It was Rick’s first show with us, and as a remote employee based in Denver, he said it was great to spend time with ScienceLogic team members who made him feel very welcome.  He definitely learned a lot at the show and met tons of new faces!

Our Senior Marketing Events Manager said, “What stood out the most for me was meeting one of our star customers who LOVES us and raved about how fast we were up and running making him money before the ink on our contract was dry.”

Rob White, our Director of Federal Sales had this to say: “I’ve been at ScienceLogic for four and a half years now and I’ve worked many trade shows during that period.   This year was completely different than those in the past.   We have so much swagger and there was a tremendous amount of intrigue from attendees.  I even had someone walk up and say ‘we can probably never afford something like this…’  I like the fact that our perception has gone from Kia to Lexus in a few short years even though we’re still a cost effective solution.”  (No offense to Kia of course!)

John Proctor, one of our Senior Product Managers, commented, “One of the things that stood out to me was that the ‘network’ is not getting any smaller.  In fact it is growing.  Whether it is the ‘internet of things’ that was discussed in one of the keynotes, the BYOD initiatives, or the visibility of IPv6, it is clear that there will be many more devices connected to the network.”

Erik Rudin, our VP of Business Development, stated, “Interoperability is at the core of what we do, as proved by our network management platform supporting InteropNet five years running.  I had several conversations at the booth about interoperability between scores of vendors to build and support the largest temporary network in the world.  That kind of requirement puts a lot of demand on your tool set to be configurable, adaptable, easy to deploy new devices, and interoperable with new technologies/API’s.  Our ‘Best of Interop’ award was a validation on our approach to interoperability.”

Erik also said, “People were excited to see the ‘Best Network Monitoring System on Earth’ and they were astounded when they saw the breadth of device coverage and platform components (ex. ticketing, event management, portal, Run Book, topology, asset).”

We clearly had a fun and productive time at Interop!  We look forward to repeating our success at Interop New York in just a few short months.

So that brings us back to the title.  Interop is all about interoperability, and our mission from the first day we started the company was to provide a platform where we could examine a heterogeneous set of technologies that you find humming along often in multiple interconnected data centers so the context of their current health could be leveraged to understand the context of real time IT operations service delivery.   For 5 years we have arrived at Interop and delivered our network management product + our engineering resources to rapidly manage InteropNet, often described as “the largest temporary network in the world.”   Each year we are faced with hundreds of operational challenges, including interfacing with brand new equipment from new vendors (sometimes early release code) that has to work together with the fluidity that is expected to carry the demanding service levels that these events require.  One new integration we did on the fly this year was with F5, another InteropNet sponsor (see one of our F5 dashboards below).  What happens in the NOC stays in the NOC, but we can self-assuredly say that what we learned this year has made our product more resilient, smarter about the latest revisions of technologies in the backbone of the network, and more confident in our approach to proactively finding problems before they impact availability.

DCM view of F5 LTM Logical Structure

Over 1000 show attendees visited our booth, it was the most jammed / packed that we have ever experienced in our previous 7 years at Interop.  The company brand is truly reaching a global audience with 20+% of the badge scans from international IT directors/organisations.

Interop 2013 was a great show for us.  Given the success of the show for the entire team, we eagerly anticipate what is in store for 2013 Interop NY this Fall.

At a recent NoVA-Py meetup, our own Brendan Mannix, Manager of Test Automation, gave a presentation on some popular test automation frameworks.  He demonstrated four separate frameworks designed to automate application testing from the unit test level all the way through stress testing, with an extended look at automated user acceptance testing.  While some meetups follow a formal presentation style, others can be open-air discussions, group hack sessions or even just grabbing a burger and cracking a few jokes about other popular programming languages.

Python is a core technology at ScienceLogic and we are proud to host the local user group meetings.  It’s in our interest to promote the continued growth of these technologies and hosting the NoVA-Py meetups  is a great way to encourage that growth while connecting us to the community.

The next meetup is tonight, May16th at 7pm, and the topic will be “Side Projects”.

To join the NoVA-Python community, sign up at http://www.meetup.com/NOVA-Python/ or follow them on Twitter @NOVAPython

Python is an interpreted, object-oriented, high-level programming language with dynamic semantics. Its high-level built in data structures, combined with dynamic typing and dynamic binding, make it very attractive for Rapid Application Development, as well as for use as a scripting or glue language to connect existing components together. Python’s simple, easy to learn syntax emphasizes readability and therefore reduces the cost of program maintenance. Python supports modules and packages, which encourages program modularity and code reuse. The Python interpreter and the extensive standard library are available in source or binary form without charge for all major platforms, and can be freely distributed.

Do Less with Less

Recent 451 addition, Tony Bishop, hosted Digital Realty in a panel that saw the wholesale data center behemoth admit to seeing smaller KW pools from customers with intensified expectations of one-stop shopping from their data center providers, and a need to be more in touch with the need of agile apps.  BBC validated the trend by exemplifying its use of Salesforce.com while simultaneously refreshing its SAP footprint and classifying the data in that system for future manipulation. At the same time, the company leverages Google (1bn video users per month) to deliver video content to its online customers. The move to a hybrid model allows it to free up internal data center space by leveraging things like SaaS to replace legacy email and CRM platforms.  However, as the company increases its complexity of suppliers, it has demanded a service catalogue and single sign-on from its vendors in order to reduce complexity to it end-users.

So does that mean everyone is trying to tie together more of the stack with facilities and app intelligence in order to do more? On the contrary, they’re trying to do less with less.  An interesting concept from CohesiveFT, that resigned the legacy mantra of doing more with less as passé.  The reason was that people want to connect fixed assets (large data center capacity) into variable assets (i.e. flexible cloud) driving the ability to do less (less burdensome work day to day) with a lot less (less responsibility and cost of underlying infrastructure in the long term).  In summation, data centers were designed for operations, while enterprise IT, leveraging the variable assets, is being designed for purpose.  As the HSBC representative emphasized, enterprises today must have a strategy around classifying services and their fulfillment models. The orientation needs to commence from business strategy down in order to get a better view of cloud and data center requirements.

How much to invest in future data centers?

As has often been discussed, the future role of the Operations Manager will inevitably turn to one of Service Orchestrator.  The data center infrastructure continues to commoditize, and open source technologies are the proof point.  Take for example, the Open Compute initiative that many believe is the signal for the end of the line for IT servers.  As Chris Swan pointed out, with such massive Technical Debt nowadays, it is becoming highly efficient to reduce your technical debt through things like Open Compute Technology.  Even companies like Google, that created Hadoop and Mapreduce for internal infrastructure efficiency, are now seeing tremendous take-up by their friends at Microsoft and Apple.   And if Google is setting the stage for emulation, bear in mind that its self-built data centers house what Gartner considers to be the world’s 4^th biggest server builder: Google itself.  The concentration then for operators (whether enterprise or MSP) is to orchestrate deployment and higher utilization higher up the stack.

Best Execution Venue

As Service Orchestrator, or broker, the IT/Ops manager, it can get incredibly confusing to choose from over 270 Cloud Infrastructure (IaaS and PaaS) providers, and a myriad of SaaS providers with growth rates faster than any other segment of technology.  And this has created a new breed of service providers, such as Gravitant. The company had a customer that tried AWS as its foray into the cloud, but was disappointed by the lack of meaningful SLA available.  In leveraging Gravitant to choose between 5 different providers, including AWS, the company found that Terremark and Savvis were more utilized due especially to backup and DR requirements.  The broker helped to mitigate the difference in charges and SLAs that can hinder, and sometimes fool, enterprise users.

It’s all about chatty apps

We live in an increasingly collaborative business-place, and these apps are no longer seen as friendly social media outlets for one’s random thoughts.  Rather, the surge in popularity that apps like Lync have brought to the business place, have also brought additional demands too.  One look at the number of PowerShell commandlets available for Lync, and there is a rapid understanding of the core function such an app can have on all business communications and collaboration alike.  More critically, these collaborative apps are driving many of today’s architectures, and seeing much faster life cycles than previous disruptive technologies.

To exemplify the point, 451’s Alan Pelz-Sharpe took a look at the history of the collaborative marketplace.  The market evolution started with document management & knowledge management systems pre-2004.  Then around 2004 email systems of Exchange and Outlook exploded globally, as pivot points for collaboration and shared documentation, calendaring, etc. This was followed by departmental shared network drives that quickly became overwhelmed. After 2008 there was the viral growth of SharePoint in the market place, followed up in 2011/12 with cloud based file sharing and mobile synch.   The point is that nothing goes away, but most industries are seeing disruption cycles getting shorter and shorter, meaning that tools must adapt to this market frame; teams that leverage tools that adapt to agility and can pivot faster than competitors in the market place will produce winning operations teams in the future.

Are you Chicken Farming?

As Joe Baguley from VMware boldly explained it: IT departments tend to look after their kittens – giving them names, taking them to the vet when they’re sick and looking after their little workloads.  AWS, on the other hand, is cultivating a culture of chicken farming – if one is sick, I kill the chicken and get on with the rest.  The issue with this approach is that whoever denounces the cloud as broken is either not trying it, or has put their kitten in the chicken farm.  The driver is that people are deploying kittens – but want chicken farm pricing.  As enterprises become serious about the cloud, they seek the resiliency and fault tolerance that is critical to modern workloads which is why VMware feels that AWS will fall the way of IBM AS/400 that started with just as much a bang in the market place.  However, the race is on, and demand will invariably drive AWS to attend to those same market needs.  In the meantime, modern day brokers are helping people choose execution venues based on internal policies that best suit different workloads.

Other companies, such as CloudSoft, enable users to move apps around at the software/app layer, when, as the CTO put it, the wind blows the wrong way, and the chicken farm smells.  But it’s the decision making prior to the portability and motility of those workloads that are evidently becoming key to any execution platform.  It’s also why we, at ScienceLogic, are aggressively building the platform for these hybrid cloud executions that many of our MSP customers themselves are leveraging alternate venues on occasion, rather than forcefully using their own infrastructure when unnecessary.  The decision to execute workloads in one cloud may be a long-term decision, but where inside that cloud (i.e. which geographic region, and best performing or most efficient cost zone) is just as imperative.  As the Director for the UK Parliament’s ICT pointed out, the innumerous metrics now from any cloud platform needed to validate the performance of a workload require a modern kind of monitoring and management tool, in order to create any assurances.

Drivers of change in our cloud environment?

So what are the workloads that are driving all of the cloud discussions today?  The tremendous growth of collaborative apps, and storage-defined networks, are both the pre-eminent drivers of cloud usage today.  From a storage perspective, a number of deficiencies to the existing systems were pointed out. For example, most legacy storage systems were not designed for multi-tenancy, nor were they developed for virtual environments.  Similarly there has been a lack of QoS in this sector since it is so difficult to predict or guarantee performance.  Finally storage is hard to scale, and once you fill it up, you either end up with multiple systems as you scale or upgrade and go through painful migration.  While physical storage costs are declining, the real cost comes in the form of managing this process, as the complexity continues its incline.

So how have MSPs dealt with these issues to date?  Often Ring-Fencing is used, with an advanced reservation of compute storage alongside application profiling to ensure the availability of resources. This can be effective but costly due to the high overheads associated with this approach.  What has instead emerged in the last 3-6 months is a software defined approach to storage. This is not necessarily new to RAID and how systems manage Cache, but there are new and interesting things starting to happen in the industry. Software defined storage in itself is not new. Vendors in this space have been making large margins off storage management software.  However, there is a set of new disruptive technologies entering the market – such as flash optimized storage, object based scale-out technologies, and SSD providers like our partners SolidFire and Intel – all driven by software defined approaches.

Some of the attributes of a Software Defined approach as per The 451 Group:

1. The software/storage runs on commodity (x86) hardware
2. A software approach allows easier scale out of storage; such as the ability of CloudFounders to detach the storage from the physical hardware in order to move data around for scalability and DR purposes
3. A unified storage layer – we are starting to replace the legacy storage silo’s speaking different protocols (such as NAS, SA, etc.) with a multi-protocol layer
4. Leveraging open source – until today open source has had little impact compared to the services space
5. API-based provisioning/management and integration, in an age of hybrid environments and toe-dipping in the cloud, to be able to point some data sets to public sources and others to internal IT storage

Ultimately things like de-duping, and having table arrays and flash drives is fast becoming the norm.  The real challenges however, are centered around scalability, common standards for portability for highly fluid applications, and the QoS currently missing in the cloud storage plays, where IOPS is fast becoming the true measurement of the effectiveness of a cloud platform.  To these points, there are increasing examples of very chatty and rapidly up and down-scaling aps, that are leveraging AWS S3 technology to such a degree that it is fast becoming the default standard in a market that is emulating its API in lieu of a formal interoperable standard.  All the more need then for a higher-level control plane from which to manage all of these technologies.

Check back tomorrow for part two of my experiences at this year’s HCTS.

1) Server Virtualization, along with increased processor capacity, has led to very dense compute models with hundreds of virtual machines on a single server which enabled these new business models. The dense compute models increased the need for redundancy and additional bandwidth (BW) in the uplinks from the server to the switch. To overcome this, Cisco created virtual Port Channels which not only increase the number of uplinks connected to a server, but also let these uplinks be split to two different physical switches. All uplinks can simultaneously be active thus increasing the network capacity as well as adding a level of redundancy.

2) VM Mobility – VMware’s vMotion enabled the ability to move a virtual machine from one physical server to another. It is this capability that really becomes important to manage work loads and provide on demand computing.  Of course there were a lot of restrictions on this concept initially due to network and storage constraints. However, vMotion set the stage for VM mobility.  In order to support VM mobility, several technologies have been created to realize this:

1. Cisco OTV – Overlay Transport Virtualization  – Cisco OTV encapsulates Layer 2 Ethernet traffic within IP packets, extending the LAN across data centers. This protocol enables vMotion across data centers. vMotion across data centers can be used for moving workloads to servers in other data centers where there is spare capacity, data center maintenance, disaster avoidance, etc.
2. LISP – Locater Identity Separator Protocol – After a vMotion occurs across data centers, this protocol enables more efficient routing of packets by separating the location information and Endpoint Identifier information.
3. TRILL – Transparent Interconnection of Lots of Links – TRILL provides some routing capabilities to layer 2 which enables large layer 2 networks to be created. Rather then having vMotion constrained to a POD within your data center, Trill allows for very large layer 2 networks allowing vMotions to operate anywhere within your data center.
4. Cisco Fabric Path is a suite of technologies based on TRILL. Cisco FabricPath uses TRILL for the L2MP capabilities and also provides support for Non FabricPath switches leveraging vPCs.
5. L2MP – Layer 2 Multipath – This is a generic term for technologies that enable multiple paths between devices at layer 2 such as TRILL. That is, currently the Spanning Tree Protocol (STP) limits the connections between devices to a single active link.  L2MP overcomes these limitations.
6. VPLS – Virtual Private LAN Server – This provides another mechanism to provide LAN connectivity across an IP/MPLS network. This can be used to extend layer 2 across data centers like OTV. VPLS does require mesh connections between sites and OTV does have some flooding advantages over VPLS.
7. SPB – Shortest Path Bridging – This is an IEEE standard (802.1aq).  SPB is an L2MP similar to TRILL. That is, SPB is a competing standard with TRILL for replacing the Spanning Tree Protocol. With Cisco’s support behind TRILL, I would bet on TRILL.

3) Server to Server Communications – According to Cisco, 76% of all traffic stays within the data center. This means a huge portion of the traffic is east-west traffic. This is a shift from data centers in the past where traffic was predominately north-south. That is, north south traffic is between a client and a server while east-west traffic is traffic between servers.  Increased east-west traffic stems from application to application communications, application to database communications, application to storage communications, and then of course the large increase due to VM Mobility. The same technology to scale the layer 2 network also provides the communications paths to support east-west traffic. That is, TRILL provides support for many links between devices effectively providing mesh connectivity while making it easy to add links to congested areas.

4) Fabric Unification, Network Convergence – Both of these terms refer to the unification of the underlying infrastructure for SANs and LANs. Instead of having completely separate networks, cabling, and equipment for SANs and LANs, converged network utilizes Ethernet for both. This greatly simplifies the data center. Key technologies introduced to support this are as follows:

1. FCoE – Fiber Channel over Ethernet. FCoE is INCITS T11 standard that specifies how Fiber Channel (FC) can be carried over Ethernet links.
2. DCE – Data Center Ethernet. DCE was a term used by Cisco prior to the standard DCB work. DCE not only included DCB-like capabilities, but also L2MP technology.
3. DCB – Data Center Bridging. DCB provides enhancements to Ethernet to support FCoE. FC requires a lossless network and Ethernet is a lossy network. DCB is a set of IEEE standards that enables FCoE. The key technologies DCB provides are as follows:
   • PFC – Priority-based Flow Control. PFC provides a mechanism to enable flow control per traffic class on an Ethernet interface.
   • ETS – Enhanced Transmission Selection –  Provides the ability to specify how much bandwidth can be  used by each class of traffic. This ensures that LAN or SAN traffic will not consume the entire BW.
   • Congestion Notification – Provides ability to notify the source that there is congestion and the source should throttle traffic.
   • DCBX – Data Center Bridging Exchange protocol. The DCBX protocol is used to discover and exchange information with DCB peers.