Skip navigation
1 2 Previous Next

IT Peer Network

16 Posts authored by: jimmywai

Industries around the world rely on new technologies to enhance their capabilities and competitive advantages. This is true for the oil and gas industry, where high-performance computing (HPC) is essential to discover new natural resources using exploration geophysics. As oil and gas exploration is competitive, risky, and very expensive, the accuracy and efficiency of seismic surveys using technology and HPC simulation is therefore critical to reduce business risk and operation cost. Imagine the business impact of reducing the timeframe of acquiring computation results of seismic simulation from 2 months to just a couple of days. I recently came across such a case with China National Offshore Oil Corporation (CNOOC).


CNOOC is China’s largest offshore oil and gas producer whose business is focused on searching for large- and medium-sized oil and gas fields offshore. To find new oil and gas resources, CNOOC relies on seismic data from offshore acquisition vessels that record high-resolution echo from sound waves bouncing off the sea floor.


As it gets deeper into its exploration activities, however, the size of CNOOC’s seismic data grows as well, with more projects and higher complexity. A single seismic project, for example, may involve over 100TB of data. CNOOC needed a more efficient, scalable, and high-performing storage system to manage and transfer large amount of data as well as ensure efficiency in collecting, recording, processing, and interpretation of seismic data.


To solve its data storage woes and enhance its oil and gas exploration capabilities, CNOOC evaluated different options before deciding to deploy a solution based on open-source software and industry-standard high-volume servers. It built large storage clusters with Intel® Enterprise Edition for Lustre* software, Intel® Xeon® processor E5-2600 v3 product family-based servers, and Intel Server Adapter X520 family, which provides 10Gb Ethernet connection for the storage cluster.


Lustre* is an open-source, parallel file system designed for HPC needs. A Lustre* file system can be scaled to multiple storage clusters with thousands of storage nodes, which makes it very suitable for HPC applications and supercomputers. The Lustre*-based solution not only met CNOOC’s storage needs but also reduced cost of performance and capacity expansion compared to its old system. As the solution comes with Intel® Manager for Lustre* software, CNOOC found it easier to facilitate Lustre* deployment, expansion, and management.


Thanks to this solution, CNOOC improved storage performance by 4.4 times while allowing it to leverage a unified storage service that simplified and centralized services for HPC projects and increased utilization rates of storage facilities and network bandwidth. Computing results that used to come out in 2 months can now be obtained in just a matter of days.


If you want to know in detail how CNOOC enhanced its storage system for improved seismic data processing, you can read the complete case study here.

Making a large investment in buying new servers and adding computing power to your data center is not a good thing, if you are not able to maximize the return on investment. Have you ever considered that the software can be the bottleneck for your data center performance? I came across an interesting case about how Tencent achieved significant results in storage system performance through software and infrastructure optimization.


Most of you are probably familiar with Tencent, one of China’s top Internet service providers. Its popular products like QQ instant messenger*and Weixin*, as well as its online games, have become household names among active online users in the country.



With the popularity of its social media products and massive user base in hundreds of millions, it is not surprising that Tencent needs to process and store lots and lots data like images, video, mail and documents created by its users. If you are a user of Tencent’s products, you are likely contributing your photos and downloading your friends’, too. To manage such needs, Tencent uses a self-developed file system, the Tencent File System* (TFS*).


Previously, Tencent utilized a traditional triple redundancy backup solution which was not an efficient solution in storage utilization. The storage media was found to be a major cost factor of TFS. As a result, Tencent decided to implement an erasure-code solution using Jerasure* open source library running on Intel® architecture-based (IA-based) servers.


As Tencent engineers validated the new solution, they noticed the computing performance was lower than the I/O throughput of the storage and network subsystems. In other words, the storage servers were not able to compute the data as fast as the I/O subsystem could move them. Adding more compute power might appear as an obvious but costly solution. Instead, Tencent used software optimization tools like Intel® VTune™ Amplifier XE and Intel® Intelligent Storage Acceleration Library to identify inefficient codes in its system and optimize them for the Intel® Xeon® processor-based server. The results were very effective and the bottleneck of system performance moved to the I/O subsystems. This was then easily addressed by Tencent when it migrated to a 10 Gigabit network using Intel® Ethernet 10 Gigabit Converged Network Adapter.


As a result of the cost-effective optimization effort, Tencent was able to get the most out of the storage system it deployed. Tencent found the optimized erasure code solution effectively reduced storage space by 60 percent and storage performance enhanced by about 20 times, while the I/O performance of TFS improved by 2.8 times. With cold data now being processed using the new TFS system, Tencent has saved significant server resources and raised the performance-price ratio for storage.


The new solution not only contributed to performance and user experience. Tencent is also saving hundreds of kilowatts of energy as it no longer needed to purchase thousands of servers to meet its storage needs.


The next time you access Tencent’s product, you now know the efforts Tencent engineers have put into improving your experience. If you are interested to know in detail how Tencent optimized its software and removed bottlenecks in its storage system, and their results, you can read the complete case study.


Have you got any interesting software optimization story to share?

This week I am hammered by a lot of spam messages on my Facebook page. I was sure those were from my lovely friends clicking on malicious links on their own page. I eventually posted a note and asking them to 'stop and think' before clicking on inviting links. A colleague responded and reminded me law #4 of Malcolm Harkins' Five Irrefutable Laws of Information Security - Users wants to click. It suddenly dawned on me that Malcolm's insight was so true. Fortunately, I'm still proud to say that I haven't fall as a victim of the malicious links and spamming my friends yet. Some of those inviting links I encountered were obviously macilious. However, some were on the border line of being legitimate. I had to hold my urge to click. Of course, some of them proved to be bad ones, too, after my friends clicked on them.


Here are the 5 laws from Malcolm. You can find them from Intel IT's whitepaper on Rethinking Information Security to Improve Business Agility.


  1. Information wants to be free - People want to talk, post, and share information
  2. Code wants to be wrong - We will never have 100 percent error-free software
  3. Services want to be on - Some background process always need to be running and can be exploited by attackers
  4. Users want to click - People naturally tend to click when they see web links, buttons, or prompts. Malware creators know this and take advantage of it.
  5. Even a security feature can be used for harm - Security tools can be exploited by attackers, just like other software. This means laws 2, 3, and 4 are also true for security capabilities.


Under the new connected, always on, and social internet age, how are you, or in fact, how should we all respond to the new paradigm and new kinds of information security risks? In my recent experience, the user is still the weakness link.

Couple weeks ago, I traveled a long way from Hong Kong to California to meet with other community managers in Intel and to attend a social media platform conference. It was a productive & very informative week. Allow me to share some quick notes and observations with you.


  • I heard the attendees at the conference were more than double compare to previous year. There were a lot of energy and passion among the attendees towards the topic. It is showing a strong trend in more enterprise adoption of social media.
  • From people I talked to, there are similar among of internal and external deployment of social media platform – internal for collaboration between employees and external for engagement with customers. Among the external implementation,  support communities seem to be one of the more popular social media implementation. There were quite a few success stories about support communities shared.
  • Executive support, endorsement and participation are key to the success of a social media program as noted by a number of speakers sharing their success stories and by a few conference attendees I spoke to.
  • A social media community requires a critical mass (a certain amount of active participants) to be successful. However, you cannot build a community and hope people will come. Marketing and getting people participating initially is key.
  • A intuitive interface is helpful to get people participate and engage. Instead of using the default user interface from your platform, customize it for your audience.
  • You need 'super users' in your community and you need to keep them engaged and motivated
  • There is still a lack of standard or best practices in measuring success of a social media program or getting actionable indicators
  • Business value of social media is still difficult to measure. A few suggestions: telling success stories; finding proxy measurements to cost reduction or revenue generation; finding a quantifiable business case


Are you seeing similar trends and challenges in your social media programs? Do you have any success stories to share?

Intel Developer Conference is coming fast on Sep 13. A few Intel IT members of this community will speak at IDF this year discussing hot topics in the IT industry and sharing how we use technology to create business value inside Intel. If you are attending the IDF next week, you are welcome to meet with them at the following sessions.


  • DianeBryant, Intel’s CIO, will be joined by several of her peers from Thomson Reuters, Genentech Informatics, Motorola and Walt Disney Studios in a CIO Panel discussing current and future trends ranging from IT consumerization, cloud computing, business intelligence (BI), security and more (SPCS002)
  • Gregg Wyant, Intel IT CTO and Dr John Pormann from Duke University will discuss the increasing challenge of large data management, data proliferation, and efficient information processing including 3D content, context awareness and more (INTS001)
  • SudipChahal and ChrisPeters will discuss Intel’s Cloud Computing strategy, investment roadmap and share our perspective on key requirements for hardware and software developers (DCCS004)
  • VirtualDave from Intel IT joins other technologists from software vendor Citrix XenDesktop, systems integrator CompuCom Systems in a panel focused on using client virtualization to create business value  (PROP001)
  • ddevette and JohnMahvi join Intel NSG to discuss Intel® SSD and share Intel's total cost of ownership model as a means to quantify the benefit of using SSDs in the enterprise (SSDS005)
  • Cindy Pickering and Ed Jimison discuss IT consumerization, a core industry trend where employees want IT to support personal devices in the enterprise (INTS002)


You can find the schedule of these sessions and plan your IDF attendance using the IDF Forum Planner. Enjoy your IDF!

At Christmas last year, I made a What I want for my work PC in 2010?to bring my own PC to work and have only one single PC for work, home and travel. That wish hasn’t come true yet. However, on the mobile device front, we do make progress.


Intel IT has been supporting a few models of smartphones to allow Intel employees accessing their emails with those gadgets. I have been using a smartphone to access work email for a few years. I've found it a great productivity tool, especially when I ride the subway to work. Previously, the number of models supported by Intel IT has been limited. With some models, employees can only get email content without attachments due to information security requirements.


In an IT newsletter to Intel employees I read last week, Intel IT announced that we are going to expand the support of smartphones by a large extent. Basically, many devices running the popular smartphone operating systems (OS) will now be supported with both email and attachments. And, it is not limited to the smartphone form factor. Tablets running those OS’s will be supported, too! I have been hearing my colleagues’ wishes that they want to access corporate email from their latest cool personal gadgets. This news certainly answers their wishes, and mine. I believe, in general, this improves employee productivity by enabling them to access corporate email with their device of choice and they can have this capability in multiple devices in their bag or at their home.


Are you accessing intranet and email from your smartphones or mobile devices? How is your IT organization supporting them? Please share your story with us at the comment field below.


If you’re interested in Intel IT’s view on Device Independent Mobility and Client Virtualization, check out our related Enabling Device-Independent Mobility with Dynamic Virtual Clients.

Intel IT runs 95 data centers worldwide with almost 443,000 square feet (41,000 square meters) collectively. These data centers house approximately 100,000 servers enabling both the operations and innovations in Intel. How is a tour inside these data centers sound to you? Sounds good? Here is an opportunity!


IT@Intel just released a series of videos that takes you through a tour of a few Intel IT’s major data centers. Not only we will show you the inside of our data centers, we will also share our approaches in facilities, compute, networking, and storage, as well as our strategy to create USD 650 million of value for Intel.


Now sit back and enjoy the ride!


If you like the Intel Server ROI Calculator, here is a good news for you! A new and improved version is coming soon!


Intel IT continued our 4 year server refresh cycle in 2009 despite the challenging economic environment. With that, We have avoided a cost of US$12M in 2009 (read more in our new 2009 Intel IT Performance Report). The learnings from Intel IT has been captured in this Intel server ROI Calculator so we can share with our peers in other IT organizations.


Here are some great new features in the coming version:

  • Ability to run the tool in both online and offline mode
  • Support for heterogeneous server environment; up to 5 server configurations can be modeled/evaluated at the same time
  • More responsive side-by-side modeling so you can see results immediately while making changes
  • More intuitive virtualization modeling that allows for adjustment of virtual machine density


Here is the teaser for the new version! Stay tuned!

Diane Bryant, our CIO, recently did an interview with ZDNet Asia. She covered a wide variety of topics including her job in Intel and hot topics among her CIO peers. Here are some key points covered. Or, head over to the article to read the full interview.


  • CIO's job at Intel IT

  • Increased dependency between business and IT

  • Tough economy... meant we needed to be much more disciplined in presenting ROI to CFO to get funding

  • Justified replacment of 20,000 servers by saving Intel US$19 million dollars

  • Popular topics among CIOs: consumerization, videoconferencing, data center efficiency

  • BYOC - Bring Your Own Computer to work

  • Top project for 2010: Windows 7 deployment


How did you see 2009 and what is your top project for 2010? Please share your thoughts below.




I don't want it! I want to bring my own Mac to work!


Consumerization, client virtualization and dynamic virtual client are buzz words that I have increasingly heard from both internal discussion here in Intel IT and sharing with other IT professionals in peer organizations. All those are seems to tell that client virtualization is going mainstream in 2010 and it is also in Gartner's top ten list as Chris What is IT Thinking?. Prehaps, it's not a coincident that we have two whitepapers (see below) on this topic within the last few weeks, and our CIO, Diane Bryant, just shared her view toward BYOC (Bring Your Own Computer) in a recent interview.


So here is my wish for 2010: I wish my Mac is my single primary PC for work, home and travel!


What's your PC wish for 2010?


Let users accessing their applications and information from any device, anywhere and anytime: Enabling Device-Independent Mobility with Dynamic Virtual Clients
Integrating Mac into our Windows environment: Using Virtualization to Integrate Mac OS X* into PC-centric Environments

I'm working with my colleagues in The Server Room to organize an Intel Data Center Live Chat on Open Port. This is the first IT@Intel Live Chat on Open Port. I'm very excited about it. Come and join us!


In the Live Chat, we will be focusing on the industry hot topics in the data centers including IT Data Center Strategies and Server Refresh. Our IT experts has recently published a white paper on Intel IT's Data Center Strategies and our approach to improve efficiency. They will attend this Live Chat to discuss our strategy, respond to your questions and listen to your views.


The Live Chat will be hosted at 10am - 12pm PST on December 8th, 2009 at The Server Room.


Please join us for the Live Chat and ask questions directly to our IT experts. At the Live Chat, we will give away a free "Energy Efficiency for Information Technology" book to all attendees. Details about the book giveaway will be annouced at the chat session.


Mark your calendar now, or download the event into your calendar directly.


Speakers at the Live Chat
Shesha Krishnapura, Senior Principal Engineer, Intel IT
Shesha is a senior principal engineer in Intel Platform and Design Capability Engineering group responsible for the development of High Performance Computing (HPC) solutions for Design, Tapeout and optimal platform foundation for Enterprise Computing. Additionally, Shesha is leading the enablement of Intel Architecture (IA) based solutions in Electronic Design Automation (EDA) industry and driving EDA application vendors to bring just-in-time IA optimized solutions for silicon design engineers.


Ananth Sankaranarayanan, Technical Program Manager, Intel IT
Ananth is a Technical Program Manager within Intel IT Core Systems Engineering. He is responsible for High Performance Computing (HPC) and Platform Engineering Programs within Intel. Ananth chairs the Technical Review Committee to recommend Servers, Storage, Backup/Recovery and Virtualization solutions for Intel Silicon Design, Office, Manufacturing and Enterprise computing environments.


Brad Ellison, Data Center Architect, Intel IT
baellison is the Data Center Architect for the IT-OPS Data Center Services team at Intel.  Brad is a former Board member of the Data Center Institute and is a charter member of the Infrastructure Executive Board Data Center Operations Council as well as ICEX Knowledge Exchange Program Data Center Excellence Practice.


Chris Peters, Strategic Marketing Manager, Intel IT
ChrisPeters is a Strategic Marketing Manager with Intel IT’s Industry Engagement Group.  Utilizing an integral knowledge of Intel IT operations, Chris works closely with strategic IT decision makers world-wide to help them optimize their IT infrastructure to deliver better business value.


Ed Groden, Product Managing Manager
egroden is a Product Marketing Manager in the Server Product Group at Intel.  Ed has been with Intel for over 9 years, and is has been involved with the Intel Xeon® 5500 processor since the inception of the program.  He works closely with OEM accounts world-wide to help integrate Intel technologies across their server and workstation products.


Ken Lloyd, Sales Engineer
K_Lloyd is a sales engineer working with the Intel and fellow traveler field in support of enterprise solutions built on Intel products and technologies. Ken has a broad background in Information Technology, systems integration, and solutions architecture.


Dave Hill, Product Marketing Engineer
dave_hill is a 15 year veteran at Intel and currently in the Data Center Group working with customers and server vendors on how to optimize their servers for best energy efficient performance.  He holds a bachelor degree in electrical engineering and has held various positions at Intel in test engineering, technical marketing and product marketing.


More on the Cash Machines...

Posted by jimmywai Oct 11, 2009

Watch Diane Bryant, Intel CIO, talks about the cash machines in data centers in this press breifing. Haven't heard about the amazing cash machines for your data centers yet?! Better check it out now: Installing Cash Machines in your Data Center

At a recent event our CIO, Diane Bryant, talked about our continued plan to replace old servers in our Data Centers ( Here is a summary of her key points:

  • Not replaceing servers could have costed Intel $19 million due to high maintenance and cooling cost
  • Our plan of refreshing old servers with Nehalem servers will save Intel $250 million over 8 years


If you are an IT manager looking at where you can find extra dollar in your IT budget to invest in new technology, new innovation and new competitive capability for your organization, this must be good news for you! Moreover, if you do nothing, you are opening a hole in your IT budget.


Here is a recent white paper and a video we published to discuss our server refresh strategy and how we are getting the cost benefit Diane Bryant shared:

Realizing Data Center Savings with an Accelerated Server Refresh Strategy


We have also developed a Server Refresh ROI estimator so you can calculater the amount of savings you can get from these cash machines:


If you ain't satisfied, here is a video showing you how to use the estimator!


Go and install those cash machines into your data centers now! 8-)

Reading from news ( today, a survey has shown that 54% workplaces block social networks completely. I'm glad to be in a company which is the 10% which allow social-network use at work so I can stay connected with my external partners and industry peers. It seems the debate on whether social media is a effective business tool or a productivity drain is still going on.


In Intel, we are embracing social media as a mean to transform collaboration in Intel. We see the opportunity out weights the potential risk. We are deploying a social media platform for our employees. You can find out more about our social media strategy from our recent white paper (Developing an Enterprise Social Computing Strategy) and the blogs from Laurie Buczek (Why Intel is investing in Social Computingand Intel's Enterprise Social Computing Strategy Revealed).


Personally, I think social media is going to repeat the history of email and instant messaging (IM) at work. Few years ago, there were skeptics about IM at work. Our CIO at that time, John Johnson, took the risk and deployed IM in Intel. Today, it's a productivity tool that I cannot live without. This morning I was troubleshooting a problem with a colleague waiting to broad a plane 16 hours away thru IM. I frequently talk to my colleagues around the world. They could be anywhere in office, at home, or on the road, when I need to connect with them. Whenever they pop up online, I can get hold of them. Without IM, life will be much more difficult and less productive.


I have been participating in a IT pilot program testing out Windows 7 in our environment. We have a Windows 7 group setup in our social media platform where we share BKM and help each other. I got workarounds from the forum for issues I ran into with the beta version of the operating system. I also contribute my findings and solutions back to the group. Together we are creating a rich knowledge base for the Windows 7 program team. The pilot users around the world were helping each other and saving each one of us a lot of time learning about the new OS, troubleshooting and finding workarounds. This is an excellent success story for social media at work. (Find out our Windows 7 experience here: The Value of PC Refresh with Microsoft Windows 7*)


What is your view of social media at work? Is your company putting up a strategy to adopt the technology?

Someone send me this Dilbert* strip yesterday. Data Center is in such high demand even Dilbert is building one. He was having trouble getting power to run the DC with air conditioning. He thought the servers would melt to a toxic bomb. I think he might be happy to know that he could actually run servers without air conditioning in a data center and he wouldn’t need to turn the DC into a museum.


Last year Don Atwood, a regional DC manager in Intel, has done a proof of concept (PoC) to challenge industry assumption in DC cooling by running a high density DC with only a air economizer and no air conditioner. In the PoC, Don ran two DC modules in parallel – one with traditional air conditioner as control; one with air economizer as the POC test. After 10 months, other than dusty servers in the POC module, there was virtually not side effect found on the 900 servers in the air economizer module. The hardware failure rate in both modules was similar, contrary to many would have believed. The biggest finding from the experiment was that we were able to reduce 67% energy consumption for DC cooling comparing with traditional data center cooling approach. Not only the reduction in energy consumption contributed to the IT sustainability programs, we also estimated using this new approach in a large 10-MW data center would save US$2.87 million annually (based on cost of $0.08 per KWH).



Have you try running your DC without air conditioning? Do you have any other innovative way in saving energy consumptions and cost in your data centers?


* Names and brands are properties of their respective owners

Filter Blog

By date: By tag: