TCP? Who cares about TCP in HPC?
More and more people, actually. With the commoditization of HPC, lots of newbie HPC users are intimidated by special, one-off, traditional HPC types of networks and opt for the simplicity and universality of Ethernet.
And it turns out that TCP doesn’t suck nearly as much as most (HPC) people think, particularly on modern servers, Ethernet fabrics, and powerful Ethernet NICs.
Read More »
Tags: HPC, mpi, TCP, UCS, VIC
Fabric-Based Infrastructure and Cisco UCS
A good segue to Fabric-Based Infrastructure is Gartner’s Magic Quadrant for Blade Servers (March 2012), by Andrew Butler and George Weiss. To fully understand the tie in with Fabric-Based Infrastructure I suggest reading the section on Cisco UCS. Their observations are important because they tie directly to the subject of this blog. You will also get a better feel for why Cisco UCS is having such rapid customer adoption worldwide.
The emphasis for Fabric-Based Infrastructure is delivering value-add functionality that enables data centers to operate more efficiently and cost effectively. A good place to start is by looking at this Gartner report by George Weiss and Donna Scott -- Fabric-Based Infrastructure Enablers and Inhibitors Through the Lens of User Experiences (April 2012). In this short research note, George and Donna go into the key drivers and reasons for the FBI architecture and the benefits that their clients have seen. My take away for the key benefits of Fabric-Based Infrastructure are:
- OpEx and CapEx savings
- Increased VM density
- Time-To-Deploy reduced from months to hours via automation and standards implementation;
- Reduce cost and complexity and improve agility;
- Improved resiliency by recreating servers and connectivity in minutes using profiles and templates
While reading about a technology innovation is helpful, actually listening to experts discuss the architecture and give their individual perspectives can be more so.
I suggest that you make time to listen to this 34 minute video with featured guest Donna Scott (a VP and Distinguished Analyst at Gartner) and Paul Perez (VP and CTO for the Data Center Business Group at Cisco Systems) -- Fabric-Based Infrastructure (FBI) in Today’s Data Center. Donna looks at the motivations and impact of customers moving to a Fabric Based Infrastructure with an eye toward what is important to adopters. Then Paul discusses Cisco UCS innovations and how they let FBI adopters achieve their goals. If you would like, you can download a podcast of the video from theCisco Analyst Reports page.
From my perspective the truly compelling part of this story is the extent to which Cisco UCS makes the promise of Fabric-Based Infrastructure a reality, while emphasizing safety, security and the risk reduction. These are critical considerations in today’s IT environment. Cisco continues to be a key innovator in data center technology and is continuing to grow from strength to strength, delivering value and benefit for your long term application solution needs.
Below is how I think a Fabric-Based Infrastructure should look. Of course I am predisposed. Cisco UCS architecture provides the ability to define and manage over 120 different server identity parameters via service profile templates, using a native tool with Roles Based Access Controls and across geographies. UCS enables you to have a distributed environment that is centrally managed. Your admins can also use CLI, custom designed tools / scripts, or third party tools as they choose to meet the needs of their current management structure.
Read More »
Tags: Cisco UCS, data center, data center management, Fabric computing, Fabric-Based Infrastructure, intelligent automation, network, Server Management, UCS, virtualization
So this is the Million Dollar Question, right? You, along with the executives sponsoring your particular VDI project wanna know: How many desktops can I run on that blade? It’s funny how such an “it depends” question becomes a benchmark for various vendors blades, including said vendor here.
Well, for the purpose of this discussion series, the goal here is not to reach some maximum number by spending hours in the lab tweaking various knobs and dials of the underlying infrastructure. The goal of this overall series is to see what happens to the number of sessions as we change various aspects of the compute: CPU Speed/Cores, Memory Speed and capacity. Our series posts are as follows:
You are Invited! If you’ve been enjoying our blog series, please join us for a free webinar discussing the VDI Missing Questions, with Doron, Shawn and myself (Jason)! Access the webinar here!
But for the purpose of this question, let’s look simply at the scaling numbers at the appropriate amount of RAM for the the VDI count we will achieve (e.g. no memory overcommit) and maximum allowed memory speed (1600MHz).
As Doron already revealed in question 1, we did find some maximum numbers in our test environment. Other than the customized Cisco ESX build on the hosts, and tuning our Windows 7 template per VMware’s View Optimization Guide for Windows 7, the VMware View 5.1.1 environment was a fairly default build out designed for simplicity of testing, not massive scale. We kept unlogged VMs in reserve like you would in the real world to facilitate the ability for users to login in quickly…yes that may affect some theoretical maximum number you could get out of the system, but again…not the goal.
And the overall test results look a little something like this:
E5-2643 Virtual Desktops
E5-2665 Virtual Desktops
As explained in Question 1, cores really do matter…but even then, surprisingly the two CPUs are neck and neck in the race until around 40 VM mark. Then the 2 vCPU desktops on the quad core CPU really take a turn for the worse:
When a VM has two (or more) vCPUs, the hypervisor must find two (or more) physical cores to plant the VM on for execution within a fairly strict timeframe to keep that VM’s multiple vCPUs in sync.
MULTIPLE vCPU VMS ARE NOT FREE!
Multiple vCPUs create a constraint that takes time for the hypervisor to sort out every time it makes a scheduling decision, not to mention you simply have more cores allocated for hypervisor to schedule for the same number of sessions: DOUBLE that of the one vCPU VM. Only way to fix this issue is with more cores.
That said: the 2 vCPU VMs continue to scale consistently on the E5-2665 with its double core count to the E5-2643. At around the 85 session mark, the even the E5-2665 can no longer provide a consistent experience with 2vCPU VDI sessions running. I’ll stop here and jump off that soap box…we’ll dig more into the multiple vCPU virtual desktop configuration in a later question (hint hint hint)…
Now let’s take a look at the more traditional VDI desktop: the 1 vCPU VM:
With the quad-core E5-2643, performance holds strong until around the 60 session mark, then latency quickly builds as the 4000ms threshold is hit at 81 sessions. But look at the trooper that the E5-2665 is though! Follow its 1 vCPU scaling line in the chart and all those cores show a very consistent latency line up to around the 100 session mark, where then it becomes somewhat less consistent to the 4000ms VSImax of 130. 130 responsive systems on a single server! I remember when it was awesome to get 15 or so systems going on a dual socket box 10 or so years ago, and we are at 10x the quantity today!
Let’s say you want to impose harsher limits to your environment. You’ve got a pool of users that are a bit more sensitive to response time than others (like your executive sponsors!). 4000ms response time may be too much and you want to halve that to 2000ms. According to our test scenario, the E5-2665 can STILL sustain around 100 sessions before the scaling becomes a bit more erratic in this workload simulation.
Logic would suggest half the response time may mean half the sessions, but that simply isn’t the case as shown here. We reach Point of Chaos (POC!) where there is very inconsistent response times and behaviors as we continue to add sessions. In other words: It does not take many more desktop sessions in a well running environment that is close to the “compute cliff” before the latency doubles and your end users are not happy. But on the plus side, and assuming storage I/O latency isn’t an issue, our testing shows that you do not need to drop that many sessions from each individual server in your cluster to rapidly recover session response time as well.
So in conclusion, the E5-2643, with its high clock speed and lower core count, is best suited for smaller deployments of less than 80 desktops per blade. The E5-2665, with its moderate clock speed and higher core count, is best suited for larger deployments of greater than 100 desktops per blade.
Next up…what is the minimum amount of normalized CPU SPEC does a virtual desktop need?
Tags: citrix, cpu, UCS, vdi, virtual desktop, virtualization, VMware, vxi
Unification is a big deal for Adena Health Systems, a medical center with multiple Ohio locations. The company’s data center was running on disparate technology that was causing outages, neglecting incoming calls, and slowing business. With some tough competition nearby, Adena’s aging infrastructure was becoming a burden; not to mention the struggle with supporting important medical applications.
Adena took action and updated its network with Cisco technology, implementing the cohesive infrastructure of the Cisco® Unified Computing System™ (UCS), based on Intel® Xeon® processors, to run their data center. In doing so, not only was the IT infrastructure unified, but the IT team and the business were also brought together to provide top-of-the-line functionality.
Alignment between the IT staff, the business, and the technology is proving successful for the company. Additionally, virtualization has decreased the physical IT footprint and is more cost effective, while enhanced communication is achieved with videoconferencing and VoIP capabilities. IT is now at the center of Adena’s business vision, allowing for the support of applications and functionality to ensure world-class patient care.
Read more about the benefits of Adena’s IT transformation here.
Tags: UCS, unified computing, voip
As we quickly approach Telework Week 2013 in March, I’ve been hearing a lot of discussion around the benefits of telework and the level of growth we have seen over the last few years in the demand for mobility.
Mobility is sometimes easier said than done, though, especially when you are a city government looking at bandwidth increases and potential infrastructure upgrades all while providing employees with a seamless virtual experience and avoiding any downtime in the process. This is exactly what the City of Jacksonville, NC did, and they are setting an example for other cities in similar situations.
The City of Jacksonville home to more than 70,000 residents and currently has more than 500 end users in its operations center that is supported by the city’s Information Technology Services (ITS) department. As the employees became more dependent on the ITS department, the department decided it needed to refresh its current infrastructure. Read More »
Tags: govtech, mobility, mobility exchange, telework, UCS, vxi