Eye on Blades Blog: Trends in Infrastructure
Get HP BladeSystem news, upcoming event information, technology trends, and product information to stay up to date with what is happening in the world of blades.

Displaying articles for: September 2009

Harnessing Horsepower: Cores, Capacity, and Code

Last week at IDF, two Intel technologists spoke about different fixes to the problem of compute capacity outpacing the typical server's ability to handle it.
For the past 5 years, x86 CPU makers have boosted performance by adding more cores within the processor.  That's enabled servers with ever-increasing CPU horsepower.   RK Hiremane (speaking on "I/O Innovations for the Enterprise Cloud") says that that I/O subsystems haven't kept pace with this processor capacity, moving the bottleneck for most applications from the CPU to the network and storage subsystems.

He gives the example of virtualized workloads.  Quad-core processors can support the compute demands for a bunch of virtual machines.  However, the typical server I/O subsystem (based on 1Gb Ethernet and SAS hard drives) gets overburdened by the I/O demand of all those virtual machines.  He predicts an immindent evolution (or revolution) in server I/O to fix this problem.

Among other things, he suggests solid-state drives (SSDs) and 10 gigabit Ethernet will be elements of that (r)evolution.  So will new virtualization techniques for network devices.   (BTW, some of the changes he predicts are already being adopted on ProLiant server blades, like embedded 10GbE controllers with "carvable" Flex-10 NICs.   Others, like solid-state drives, are now being widely adopted by many server makers.)

Hold on, said Anwar Ghuloum.  The revolution that's needed is actually in programming, not hardware.   There are still processor bottlenecks holding back performance; they stem from not making the shift in software to parallelism that x86 multi-core requires.

He cites five challenges to mastering parallel programming for x86 multi-core:
* Learning Curve (programmer skill sets)
* Readability (ability for one programmer to read & maintain other programmer's parallel code)
* Correctness (ability to prove a parallel algorithm generates the right results)
* Scalability (ability to scale beyond 2 and 4 cores to 16+)
* Portability (ability to run code on multiple processor families)

Anwar showed off one upcoming C++ library called Ct from RapidMind (now part of Intel) that's being built to help programmers solve these challenges.  (Intel has a Beta program for this software, if you're interested.)

To me, it's obvious that the "solution" is a mix of both.  Server I/O subsystems must (and are) improving, and ISVs are getting better at porting applications to scale with core count.

Tips to Reduce Processor Latency

For some of financial and data-acquisition applications, it's more important to finish one calculation super-fast than a bunch of calculations slightly slower.  There's a group of HPC apps with a similar requirement:  two identical instructions need to have precisely the same latency, every time they're executed.

Real-Time Operating Systems (RTOS) can help address these two scenarios.  These OSes address latency in a number of ways; for example, by ditching device-polling and background cleanup tasks that that standard OS's normally do.

However, some features of modern industry-standard servers can hurt low- and consistant-latency computing.  For example, low-power processor modes might save power, but any such processor throttling can increase latency.  Another example would be management routines that consume CPU cycles, such as routines built into the BIOS of ProLiant server blades that occasionally use CPU cycles to track resource utilization and monitor correctable memory errors in the memory controller.

If you face these situations and have already gone with an RTOS, HP's got some settings in our RBSU (ROM BIOS Setup Utility) that can offer additional help.

Load up RBSU (accessed by pressing F9 while the system is booting), and change the following settings:
1) Set "ProLiant Power Regulator Mode" to "Static High Mode".
2) Disable processor c-state support. 
3) If you are running an application that is single-threaded, set "Processor Core Disable" to "One Core Enabled".
4) On Intel Xeon 5500-based servers (like the BL460c G6), disable "QPI Power Management", and ensure "Intel Turbo Boost Technology" is set to "Enabled".

If you want to go even further, there's a way to disable some of those periodic BIOS checks on processor utilization and correctable errors. For most G5 and G6 server blades, HP has a tool called conrep (provided with the Smart Start Scripting Tool Kit) that let you control these settings.

In the BL280c G6, BL460c G6, and BL490c G6, you can also disable those things straight from RBSU.  Hit "Control-A" within the RBSU, and some additional options will appear in the
"Service Options" menu.

Same challenges, same complexities, tighter budget, smaller staff - Midsize businesses virtualization benefits

Midsize companies have many of the same problems as large enterprises. Their servers are underutilized, their storage systems are difficult to manage, their disaster recovery plans are inadequate, and power and cooling costs are killing them. They lack the space to expand their IT environment, the budget to buy new hardware, and the resources to deploy new technologies.

Unlike enterprise companies, midsize businesses are much less likely to utilize virtualization technology to address those challenges. Unfortunately, they often perceive virtualization to be a complex, costly solution— one that’s great for “the big guys” with specialized IT staff, but out of reach for the average business with a tight budget. As a result, they miss out on the tremendous benefits virtualization has to offer.

However, there are several critical business and technology points that midmarket customers can benefit from when utilizing virtualization in their environments:
• Optimize the Use of Technology Assets: HP and VMware virtualization solutions results in fewer assets to manage, power, store, and buy—enabling midmarket customers to increase utilization rates for x86 servers from 5–15% up to 60–80% while simplifying IT operations.
• Simplify Asset Management and Administration: VMware Infrastructure is easy to deploy, use, and manage. It provides IT professionals with the freedom to be released from menial tasks, enabling them to take a more strategic role in the enterprise.
• Protect Data and IT Environments: Through advanced technologies, HP and VMware offer business continuity solutions for improved disaster recovery, higher levels of availability and simplified backup and recovery. Provide better levels of security, improve responsiveness, and increase the resilience of your IT environment.

HP and VMware Virtualization solutions are ideal for midmarket
Check out the reduced complexities of virtualization by integrating industry-leading VMware ESXi into the ProLiant platform. Virtualization software is IN THE BOX without any need to need to buy extra hardware or software.

Chris Purcell

More Impressions - VWworld 2009 - Putting virtualization to work to change the economics of IT

Day 2. I sat through two sessions this morning, one with Dr. Stephen Herrod on the future of Virtualization: From the Mobile Phone to the and the Cloud and the other Conquering costs and complexities in a virtualized environment: Research and case studies with to HP guys and a lady from IDC. Both were very interesting sessions.

The both sessions touched on a tectonic shift that has already started to happen within the industry as datacenters start to virtualize. This will have a couple of significant impacts. Firstly, it is predicted by 2012 the number of virtual servers will have outgrown physical servers, and secondly, the overall growth of servers will start to flatten. In 2005, 7 million servers were shipped, 1 million of them were virtual servers. By 2012 the predicted shipped amount of servers will be 9 million physical and 16 million virtual. It has been an amazing growth, but that will taper off as the volume will be so great.

One of the biggest issues that need to be address is the virtualization management gap between virtual and logical servers. Steve Herrod is quick to point out the growing number of VM tools becoming available and how easy it will be to manage all assets from servers to desktops to iPhones. But the underlining fabric of the datacenters needs to rethought to accommodate this shift. Right practice tools need to be able to manage the infrastructure and high stakes are placed around disaster recovery as datacenters consolidate. These are no longer an option.

I really enjoyed the section in the first session where Steve Herrod demonstrated access to his personal desktop via a iPhone app called PocketCloud from Wyse. This demonstrated a path to manage the data center via a handheld devise which is almost within our grasp. There was also a Visa app for the iPhone were you can track you visa expenses real time, and then when you run out of cash, Google maps is available to show you where the nearest ATM machine is. Too funny.

But all this is very dependant on the stability of the infrastructure and the ability of removing all the five “9” and replacing them with “0”, getting power and cooling costs right down and the ability of moving into the Cloud. So how long will it take s us to get fully integrated into the cloud? IDC thinks we are still 10 to 15 years away and we still have a very interesting journey to get there. I will enjoy this journey.

This is my last full day today, but I will be publishing a few more articles before I leave San Francisco tomorrow.

And if you want to keep up with up-to-date VMworld 2009 news, don’t forget to follow us on Twitter @HPStorageGuy, @BladeNews @ProLiant

Chris Purcell


Labels: VMworld 2009

VWworld 2009 Impressions - 3 things I learnt today

Okay, I have to admit I learnt a lot more than just 3 things today at the VMworld 2009 event. Outside  of making the mistake of visiting San Francisco in the middle of August thinking that it would be as warm as it is back home in Texas, the trip is going well and the VMworld event has been really interesting.

I spent the first session of the morning attending the Paul Martiz’s open address. Top items:
• 96% of all fortune 100 companies use VMware software – and that if you want a free ticket to next years conference successful sell VMware software into one of these companies
• Over 70% of current IT budget (in most companies) is spent on keeping the “lights on”. Less than 30% is spent on innovation.
• Biggest realization that companies get after virtualizing is that it frees up resources to do other things and speeds up adding in new capacity
• Virtualization = easier to manage infrastructure
• The one biggest expense in the datacenter is storage – even in a poor economy data is not decrease, it continues to grow
• vSphere provides full fault tolerance

During Ann Livermore’s address at 9.30, the top items were:
• HP is the leader in VMware solutions
• 3 big trends – DataCenter transformation, infrastructure expansion and everything being provided as a service
• Between 2005 and 2012 there wil be a 10X growth  in Virtual machines – all of which needs to be planned and managed
• That virtualizing will increase productivity
• Mark Potter cited many consolidation examples of where virtualizing had decreased the overall volume of servers and had provide more control of the virtualized infrastructure.
• And most importantly, stop virtualizing servers, start virtualizing infrastructure

More tomorrow, so stay tuned.
And if you want to keep up with up-to-date VMworld 2009 news, don’t forget to follow us on Twitter @HPStorageGuy, @BladeNews @ProLiant 

Chris Purcell 9/1

VWworld 2009 Impressions - Growing challenges for IT – clear opportunities with virtualization

Data centers around the world are wrestling with a common set of challenges. Over half are not optimized for agility. Most are not agile enough to easily respond to changing client demands or unexpected execution problems. In many data centers, a good portion of the IT budget is still trapped in maintenance and operations, due to aging equipment and a reliance on labor intensive manual process. Many data centers are power and space constrained, with a sizable part of the IT budget going to data center power and cooling, which now exceeds the cost of new systems.

So after the first full day at VMworld, I heard this message over and over again – it is really the underlining theme of the show.

In Paul Maritz’s opening session he goes into detail that over 70% of IT budget is spent on “Keeping the lights on” and less then 30% goes towards “innovation”, and that the biggest costs within the datacenter is storage – clearly an element that does not diminish even in a down turned economy.

These costly problems are aggravated by legacy hardware that is locked into rigid infrastructure silos, each devoted to a particular application or business service. Resources, which can’t be easily shared, are chronically under-utilized. While some applications run short on processing power, storage capacity, and network bandwidth, other applications have far more resources than they need.

In a rigid environment like this, IT isn’t positioned to quickly provision or shift resources to meet the changing needs of the business, and everyone is short on time. IT staffers invariably have to devote too much of their attention to maintaining systems, meeting compliance requirements, and shoring up defenses against security threats—rather than focusing on innovation that increases business value. That’s a problem because now, more than ever before, IT is expected to quickly deliver new services that drive business growth, while working to cut costs. How do you meet these expectations? In short, you need new, more advanced technologies and services that break down technology silos so you can easily pool, share, and allocate resources. What’s more, these new technologies and services need to be easier to buy, quicker to deploy, and easier to manage—so you can extract greater returns from your investments.

All the while, IT organization needs to make hard decisions about compute platform choices for business services. Should you build and manage your own in-house data centers? Use a traditional outsourced service provider? Tap into cloud-related services? Or deploy a hybrid of all three?

This much is certain: The current economic environment is changing how businesses operate—and demanding more from IT. The economics of IT need to change.

Judging by the VMworld attendance of over 12,000 plus people, virtualizing infrastructure is clearly a hot topic to a lot of businesses

And if you want to keep up with up-to-date VMworld 2009 news, don’t forget to follow us on Twitter @HPStorageGuy, @BladeNews @ProLiant   

Chris Purcell 9/1


Labels: ProLiant| VMworld

One tool to manage both virtual and physical infrastructures

Introducing HP Insight Control suite for VMware vCenter™ Server: Because Knowledge is Power

By Lorre Bumgardner (Product Manager, Insight Control for VMware vCenter Server)

Last week at VMworld, HP and VMware announced HP Insight Control for VMware vCenter™ Server, the first product of its kind, to help you manage physical and virtual assets through a single, integrated tool.  HP Insight Control for VMware vCenter™ Server delivers powerful HP hardware management capabilities to virtualization administrators.

We’ve heard VMware users asking for tighter integration between virtual and physical management, a simpler user experience and greater control over your entire infrastructure in a simple, familiar interface.  The answer is Insight Control for VMware vCenter™ Server.  Streamline day-to-day monitoring and management tasks of your physical and virtual infrastructure with comprehensive monitoring, remote control and power optimization directly from the vCenter™ Server console. In addition, Insight Control delivers robust deployment capabilities that make it easy to rapidly deploy VMware ESX hosts.

From the VMware vCenter™ console, Insight Control lets you:

  • Take care of issues before they become problems: Receive pre-failure alerts when key hardware components and subsystems reach pre-defined thresholds.  With this insight, you can proactively address potential hardware failures before they impact virtual machine performance.

  • Remotely control your infrastructure: Remotely manage and troubleshoot HP ProLiant and BladeSystem servers using HP Integrated Lights-Out Advanced capabilities directly from the vCenter™ console.

  • Optimize power utilization: Get the most out of your existing power envelope by comprehending and proactively managing power for hosts and pools of virtual machines across hosts

All of this adds up to lower costs and improved administrator efficiency.

Are you interested? Let us know your take on this introduction… will it improve your management? Do you have any questions for the Insight Control team?

For more info on HP Insight Control for VMware vCenter™ Server, go to www.hp.com/go/ICEvCenter.  






Showing results for 
Search instead for 
Do you mean 
Follow Us

About the Author(s)
  • More than 25 years in the IT industry developing and managing marketing programs. Focused in emerging technologies like Virtualization, cloud and big data.
  • I am a member of the Enterprise Group Global Marketing team blogging on topics of interest for HP Servers. Check out blog posts on all four Server blog sites-Reality Check, The Eye on Blades, Mission Critical Computing and Hyperscale Computing- for exciting news on the future of compute.
  • I work within EMEA HP Servers Central Team as a launch manager for new products and general communications manager for EMEA HP Server specific information. I also tweet @ServerSavvyElla
  • HP Servers, Converged Infrastructure, Converged Systems and ExpertOne
  • WW responsibility for development of ROI and TCO tools for the entire ISS portfolio. Technical expertise with a financial spin to help IT show the business value of their projects.
  • I am a member of the HP BladeSystem Portfolio Marketing team, so my posts will focus on all things blades and blade infrastructure. Enjoy!
  • Luke Oda is a member of the HP's BCS Marketing team. With a primary focus on marketing programs that support HP's BCS portfolio. His interests include all things mission-critical and the continuing innovation that HP demonstrates across the globe.
  • Global Marketing Manager with 15 years experience in the high-tech industry.
  • 20 years of marketing experience in semiconductors, networking and servers. Focused on HP BladeSystem networking supporting Virtual Connect, interconnects and network adapters.
  • Working with HP BladeSystem.
  • Greetings! I am on the HP Enterprise Group marketing team. Topics I am interested in include Converged Infrastructure, Converged Systems and Management, and HP BladeSystem.
The opinions expressed above are the personal opinions of the authors, not of HP. By using this site, you accept the Terms of Use and Rules of Participation.