Sustainability ? Awareness of Energy Use in the Office

First, Happy Earth Day Everyone!

A couple of weeks ago, I told you about a small proof of concept we conducting to measure energy use in the office environment and to then use that established baseline to test different energy saving methods.Today, IHow about you?-Mike

Embracing Change - an Intel Team's Adoption of Agile Practices

OK, like I promised in my previous blog, here is my first post on our team's Agile adoption journey with many more posts on best practices to come in the next few weeks.

Our team is a cross-geo team - made up of a development / engineering team in Bangalore and business stakeholders / program managers in the US. Even though we were delivering good, on-time releases using the Waterfall Methodology, we felt that the team could do better. The quality of our releases started to wane with lots of post-production defects. There were heroic efforts from some individuals and clearly they were leading, but a good number of folks in the team were followers. Working late nights and weekends was a commonly observed behavior. Our stakeholders wanted more frequent releases (at least 2 or more per quarter). Our attrition rates in the team were also quite high - touching 30% in 2006.

Clearly we needed a better approach for Software Development. to resolve some of these challenges Agile was starting to get quite popular in 2007. As a management team, we talked about introducing Agile in our team. We shared our analysis of the situation with the team and also our rationale for introducing Agile. There were many skeptics in the team, there were also a few that believed in it and were willing to give it a try. We had driven the Agile adoption purely top-down. There are always risks associated with such an approach. Firstly, the team does not accept the changes easily, they start resisting it. Secondly, it was quite easy to go back to old ways of operating.

This is when we decided to get some external consulation help and hired some Agile coaches. This approach worked well for us. The coaches trained us on the methodology, showed us how its done and practically hand-held us through the adoption. They were with us for almost 6 months and once we gained some confidence in our ability to do it ourselves, we let go of them and started practicing it ourselves.

This was difficult in the beginning, but then we adapted to it quite well. We stuck to the approach. We created metrics to measure our progress. We created surveys to measure the maturity of the Agile practices. We relegiously used retrospectives as a way to improve ourselves. We took "what can go better" feedbacks from the team very seriously, assigned AR's for the team members to go resolve that and made small course corrections.

OK, but what about the results.... The results we achieved through this process was very good. Quality improved significantly. Productivity of the team improved 3X. Our stakeholders are happier. Team is happy (0% attrition and no overtime worked in the last 2 quarters).

We have more to go in this journey, but its immensely gratifying to see that teams can achieve lot of great things, if they work hard at it and stick to a strategy.

I'll share some best practices we've learnt through our Adoption of Agile in my future blogs. Stay Tuned !!

Raj

When Xeon 5500 series ?Nehalem? servers are just not enough?

Sure, Intel® Xeon® 5500 Series Processors represent a quantum leap forward in terms of both performance and energy efficiency. That has been proven in a number of test results and reviews.  But for your back-end data demanding enterprise app deployments, large scale server consolidation or virtualization of business critical applications, Intel® Xeon 7400 series processors offer outstanding performance and performance per watt in 4-socket servers. So, which platform do you choose, especially when this decision is likely going to be the key determining factor for capital savings, efficiency and TCO for your datacenter infrastructure? Well, you?re read a lot about Xeon 5500 series Nehalem servers over the last few weeks.  Let me share with you some reasons to consider a Xeon 7400 series 4-socket server when you are presented with the choice between Intel?s two best of breed products for virtualization.

4 Socket and above servers (Xeon 7400) are purpose built ? just like a large truck: They?re purpose built for your most data demanding enterprise applications like database and ERP, and for large scale server consolidation using virtualization. Large Trucks are also purpose built.  They?re purpose built for hauling large loads over long distances.   Now, you don?t buy a large truck to commute to work in.  You also don?t take your everyday commuter and attempt to haul large loads with it, because if you did you would be significantly undersized (you?ve all seen those cars on the road with rear tires about ready to pop under the weight of a palette of heavy goods tied on top). 

More Resources Matter for 4 Socket MP Workloads: The apps/workloads listed above benefit from the expanded feature set associated with 4 Socket Xeon 7400 based servers: more processors (4 vs. 2), more cores (24 vs. 8), more memory (32 dimms vs. 18 dimms), more I/O capacity (7 slots vs. 4) and larger cache (16MB vs. 8MB).  These features and what they enable are why MP Server buying patterns have remained stable with IT for the last 5 years and will continue to be stable for the foreseeable future according to IDC. 

But in today?s economy there may be MP customers out there that will want to push the envelope and attempt to deploy lesser expensive 2S systems for traditional 4S solutions. Would doing so pencil out from a TCO perspective? Let?s take a look at two Virtualization usage examples and find out.

Large Scale Server Consolidation: Where almost 2x the memory matters.

In this scenario, IT Manager is dealing with numerous corporate acquisitions across the country prior to the economic downturn, with servers that now need to be consolidated to cut costs quickly.  Goal is to convert 1000 older underutilized 2S servers.  He (she) converts these to 1000 VMs and transfers them electronically to the central Data Center.   He determines that these infrastructure apps when consolidated generally run into memory constraints before they run into processor constraints, so for his candidate solutions he compares a 4 Socket Server with Xeon X7460 processors vs. a new 2 Socket server with Xeon X5570 processors.   He fully loads both systems with 4GB dimms (128GB on 4S vs. 72GB on 2S), and assigns 4GBs memory for each VM deployed (enabling 32VMs per server resulting in 31 new 4S servers vs. 18 VMs per server resulting in 56 new 2S Servers.)

Now, he only propagates the 4S Solution with 2 Xeon 7400 Processors, which allows the IT manager to still use all 128GB of memory on the 4S Servers while paying lower VMWare licensing costs.  Price these systems out on Dell, HP, IBM?s or Sun?s website, and the Xeon X7460 servers will be in the $15k-$20k range vs. the Xeon X5570 based servers will be in the $10k-$12k range (i.e. roughly 1.5x higher for 4S vs. 2S server).  Add VMWare license costs, power/cooling, LAN/SAN cabling, and system maintenance costs and you?ll see the 4S solutions offer a lower cost per VM.

Virtualizing Business Critical Workloads: Where 3x the Processor Cores matter.

In the previous example, we were looking to maximize consolidation ratios.  In this example, we?re looking to achieve predictable high performance for a business critical app.  Solutions like ERP that are put into a virtualized environment perform best when run without oversubscription, where you set the same number of virtual CPUs to equal the number of physical cores available on the platform.  This helps deliver relatively more predictable performance for all VMs and is the way that IT@Intel intends to deploy ERP in a virtualized environment as they begin to test this moving forward (read more about this in the new whitepaper).  In this example, we?ll convert ~100 non-production ERP instances (i.e. the instances used for QA, Dev, and Production break fix).  We?ll assign 2 virtual CPUs and 8GB memory for each instance.  The four-socket Xeon 7400 processor based systems (with 96GB memory) will have a total of 24 cores and will have a list price of about $25k.  This allows us to run 12 Virtual Machines without oversubscription on the MP Servers and enables 100 ERP instances to be consolidated down to about 8 MP (4 Socket) servers.  Since the Xeon 5500 based Servers just have 8-cores, the IT manager decides to avoid oversubscription and deploys 4 virtual machines ? consolidating down to 25 DP (2 Socket) servers with 32GB Memory and a list price of about $8k per server.  Include the costs of the hardware, VMware ESX license costs, power/cooling, cabling, and Server maintenance ? the MP (4 Socket) solution here would also offer a lower cost/vm than the Xeon 5500 based DP (2 Socket) solution due to having 3x the processor cores on 4 Socket.

When you are deploying your most data demanding enterprise applications and implementing large scale server consolidation, Xeon 7400 based servers represent a very intelligent choice. 

Let me know what you think.

bryce

Sales Brief or Product Brief ? What is the Difference? ? Do You Have A Preference?

We all live in the information age and are bombarded constantly by more message and information than we can realistically consume (sorry for adding with this blog). I love Facebook, am involved in Linked-In and am exploring (as a newbie) Twitter. These forums and tools are really cool and if I wanted to, I could spend my life playing with them.  Last week, someone described twitter as a river of information and guided me to jump in and paddle downstream, not upstream.

So, that is the input I?m looking for on this blog. I just posted 2 documents as resources in the Server Roomabout the new Intel Xeon processor 5500 (Nehalem) product.

ð        Sales Brief? 2 page brief focused on the benefits of purchasing new servers

ð        Product Brief? 12 page brief focused on all the usages, technology and benefits inside the new servers

I?m told that people don?t like ?sales? briefs.  Personally I like short and sweet, but give me the tools to dig into more detail if I want. 

? that?s just my style ? What?s Yours?

1) Which do you like more, if you have a preference?

ð        Sales Brief

ð        Product Brief

2) What do you do?

o        IT ? I deploy server technology to benefit my business.

o        Sales ? I sell or re-sell technology to IT and business owners.

o        Developer ? I design and build servers or use them to design software products.

The time is right now to move from risc

I'll be up front, I really don't know what Brittany Spears, Miley Cyrus or Susan Boyle would say about moving from RISC to the Xeon 5500 processor!. What I can share is the feedback that I'm getting direct from Customers. I'm currently out on the road and have got some real feedback direct from Customers on why they are looking at migrating their solutions from RISC  processors to Xeon processors.

Over the past couple of days I have had the opportunity to meet directly with individual Customers and hosted a roundtable with several Customers to discuss their plans to replace their RISC based infrastructure. The conversation has been very open and frank and has not been about 'should I move' but more focused on 'how do I make the move'. As could be expected the down economy is placing big taxes on the ability of IT organizations to support their business units need for organic growth in a flat to down IT spending environment. A big priority for most of the Customers that I spoke with is how to reduce their overall TCO while still meeting the increased demands being placed on IT by their business Partners. Most of the Customers are already engaged in active projects to assess moving from RISC or are building their plans to make this migration.During the roundtable I had opportunity to share the latest Xeon 5500 processor performance comparisons Vs the main SPARC and POWER based solutions out there. There was great rejoicing and joy (ok I'm taking poetic license here) in the roundtable when we share some of the results that we highlighted when we launched the Xeon 5500 processor just over 3 weeks ago. So I want to spread the joy and let you read for yourself the performance and price performance benefits.We compared the Xeon 5570 processor Vs the top UltraSPARCT2+ in a 2 socket configuration. We took best published results on spec.org and sap (so no funny games at play). The results comparing best UltraSPARCT2+ vs best Xeon 5500 with 1 taken as baseline for SPARC redults were amazing- 20% better on SAP-SD- 62% better java performance for Specjbb2005- 69%better for integer performance SPECIntrate-2006- 75% better for floating point performance SPECfprate-2006But the best bit was the cost competitiveness of the Xeon 5500 solutions. Comparing both solutions with 32GB memory, the Xeon 5500 based solutions are offered at approx $11,000 whereas the UltraSPARCT2+ is at $36,000.Compared the Xeon 5570 processor Vs the top POWER6 in a 2 socket configuration gave even more staggering results. At the roundtable today Customers were amazed. They keep hearing that POWER 6 has leading performance and more GHz so better performance. Right?. Wrong is the answer and I noticed many customers scribbling down the comparisons. Again taking 1 as baseline for POWER results- 150% better on SAP-SD- 190% better java performance for Specjbb2005- 126%better for integer performance SPECIntrate-2006- 90%better for floating point performance SPECfprate-2006But the best bit was the cost competitiveness of the Xeon 5500 solutions. Comparing both solutions with 32GB memory, the Xeon 5500 based solutions are 92% less expensive than equivalent POWER 6 offerings.I only shared the specific comparisons vs RISC and have not gone into the architectural advancements of the Xeon 500 processor and how it addresses real business needs that have been flagged to us. There have been lots of other blogs out in cyberspace over the last few weeks on improvements in IO, low latency etc. so you don't need my 2 cents.I think now is the time to make the move from RISC, what do you think?

Just Published: Intel vPro Technology Heartbeat newsletter

The latest newsletter is now available - you can check it out here: http://eepurl.com/1I2

This newsletter is a good resource for documentation updates, known issues, new software downloads, and more. We have plans in the works for more training webinars - this is a good way to keep up on them!Here's the subscription link if you'd like to see this newsletter in your inbox (it's a bi-weekly newsletter): http://eepurl.com/Tqw

Shower Curtains in the Datacenter? - Live from the Blades System Insight Event

I've been in Las Vegas this week for the Blades Systems Insight event talking about data center transformation and data center efficiency (no white tiger sightings...just technology this week in Vegas).  This event draws attendees who are deploying high density compute platforms in their data centers and dealing with the power and cooling challenges that come along with these environments. So I was excited to share some of Intel's thoughts on power and cooling optimization beyond pure system refresh.  If you read the blogs on the server room you know plenty about the compelling financial benefits associated with refresh...and if you haven't seen this yet check out my friend Chris Peters' blog here.

But back to the show and the shower curtains...If you dip a bit deeper into the challenge of data center efficiency, three primary focus areas emerge:Power: The underlying power cabling and infrastructure into your datacenter.Cooling: The HVAC systems, fans, and ducting installed to remove heat from your datacenter and let you avoid thermal environments that make Las Vegas feel chilly.Compute: Server, network and storage gear that drive business producitivity for your organization.At the Blades event we were discussing the impact of high density environments to this fragile ecosystem.#1 Warmer datacenters.#2 Cool aisle containment: This is a pretty simple concept - placing barriers to control cool air and confining it to the area where servers need it.#3 Ambient air cooling: Even in Las Vegas datacenters are utilizing outside/filtered ambient air economizers instead of their chillers to deliver cooled air at least part of the year.#4 Liquid cooled cabinets - think of these essentially as a good Sub-Zero for the datacenter and especially applicable for the high density environments that we were focused on at the blade conference.I will be back to you on the power and compute vectors next...in the meantime I'd love to hear if your datacenter has implemented any of these approaches and any results you've been able to measure.

Parallel Programming Talk - Paul Steinberg Talks Teaching Parallel Programming

On today's episode of Parallel Programming Talk Clay Breshears and I sat face to face for only the 2nd time during our 27 show run.

Download the MP3 of the show.

News:

Intel® Parallel Studio and is on schedule to go live in Mid 2009. Learn more and download the beta today.

Learn more about Intel Xeon 5500 Series processors.Nehalem architecture for servers and workstations is now here.  Leading members of the software industry talk about the new Intel Xeon 5500 Series processors and how developers / data center managers can take advantage of its performance and efficiency.

After the success of the 2008 Threading Challenge, and feedback from many of you who participated, we have decided to bring it back again this year.   We know you love a good puzzle; so our multithreading experts have agreed to pull together some new and interesting brain teasers to give you a chance to flex your threading skills in 2009.The first problem, Radix Sort, is due April 24, 2009. The second challenge, 3SAT, went live April 20th. You will have until May 8, 2009 to submit your entry.  Please see official rules for more information or visit the forum for this problem to get your questions answered.

Check out the Intel Concurrency Checker 2.1 demo video by Gaston Hillar. The free utility allows anyone to test multicore optimizations in an executable. Download Intel® Concurrency Checker 2.1.

Enter the Level Up 2009, Intel® Visual Adrenaline Game Demo Challenge.
A worldwide competition providing developer?s greater exposure to the gaming community. Submissions from professional, aspiring and student game developers will be accepted until July 1, 2009. The games will be judged in three categories: Best Threaded Game, Best Game-on-the-Go, and Best Game Optimized for Intel® Graphics. Valuable prizes will be awarded to help jump start the success of the winning games.

Quick Overview of the Parallel Programming & Multi-Core Community
I put together a three minute video to give new visitors to the Intel Software Network some quick pointers to the most useful features, benefits and resources of the Parallel Programming & Multi-Core Community. Please take a moment to watch the video and if you want a quick link I have included them the text with at the bottom.

Send in your questions.
The first Tuesday of every month we will pick a listener question and do our best to provide an answer. Send in your questions to parallelprogrammingtalk@intel.com and listen in on May 5th to see if you question is the one selected.

Today's Show:

On today's episode of Parallel Programming talk we will be discussing the Teaching Parallel Programming with Paul Steinberg. Paul is the community manager for the Intel Software Network's Academic Community. Paul and Professor Tom Murphy of Contra Costa College will be launching a sister show on April 21st at 10:00AM PST.

Professor Tom Murphy of Contra Costa College and Paul will be co-hosting Teach Parallel. Please attend online or catch-up with the podcast after the event.  Our first guest will be Dr. Dan Garcia of the University of California Berkeley.  He will be talking about how he was able to integrate parallelism throughout the curriculum.  Anyone who knows Dan will tell you that he is always a fascinating and exciting speaker.

On the Next Show:

On our next show on April 28th we'll be talking to Professor Marc Snir. Professor Snir is  Michael Faiman and Saburo Muroga Professor in the Department of Computer Science at the University of Illinois at Urbana-Champaign and has a courtesy appointment in the Graduate School of Library and Information Science.  He currently pursues research in parallel computing. He is PI for the software of the petascale Blue Waters system and co-director of the Intel and Microsoft funded Universal Parallel Computing Research Center (UPCRC).

Don't forget to send in your questions.

Connecting with Teach Parallel

P-MAPS: An on-demand hardware-rooted system for protecting critical applications

Since our last virtual discussion (June 2008), malware attacks continue to rise, and more so, attacks have continued to become stealthy and targeted. We have completed a key milestone for our software protection research last month; we created a research prototype of a hardware-assisted application protection capability called “Processor-Measured Application Protection Service (P-MAPS)”. The goal of this work has been to significantly reduce the Trusted Computing Base (TCB) from a full Operating System to a substantially smaller P-MAPS layer to improve the runtime security of critical applications running within the OS. The main contributions of our work are the on-demand trusted instantiation of P-MAPS and the use of P-MAPS to protect applications without interrupting the natural operation of the application or the Operating System. With P-MAPS enabled on a platform, day-0 attacks and attempts by unknown malware to attack critical applications can be mitigated.

Dynamic Software Application Protection white paper View .pdf

We built the P-MAPS layer to be OS-agnostic; an untrusted OS-specific service is used on the platform that runs within a commodity OS. Initially, the OS is in the TCB - the P-MAPS launch put the platform in a reduced-TCB state (with the OS outside the TCB). The OS-specific P-MAPS service can be triggered by the user/OS launching an application that uses P-MAPS for protection, and can be torn-down securely when not needed by any protected applications. The P-MAPS TCB consists of the CPU, the verified chipset and platform firmware. We use Intel® TXT to measure the P-MAPS layer which allows the P-MAPS Core to be independent of the chipset. The chipset specific code is contained in the Authenticated Chipset Module (or ACM) that is signed by Intel. The processor (via the Intel® TXT GETSEC instruction set) verifies the ACM. Additionally, the ACM can verify the P-MAPS measurement against a Launch Control Policy embedded on the platform. This approach protects the user for malicious software that may try to spoof the P-MAPS layer or try to deny P-MAPS execution. To protect the applications after the P-MAPS layer has been launched in a trusted manner, we use Intel® VT capabilities. Note that with P-MAPS active, we have moved the OS execution into “guest” mode. The applications that “register” with the P-MAPS are subject to an in-memory authentication process after which they are protected as was described in our previous post. Protected applications can continue executing within the OS without any disturbance to the OSes operation or the operation of other unprotected applications.

We have written several applications that use the P-MAPS to provide three core security properties: 1. Isolation of the application’s runtime memory from other software on the platform, 2. Encapsulation of the application data memory such that only code in the measured application pages can access the data. 3. Prevention of circumvention of any function entry-points exposed in the application code. A protected application typically involves handling of secret data that is provisioned by a Provisioning Entity (Server) in the network. We have built P-MAPS such that the hardware can authenticate the P-MAPS core when it interacts with the platform root of trust (in our case, a Trusted Platform Module or TPM) which can then be used to provide hardware-derived quotes to a trusted remote entity. The TPM quotes are used by the remote entity to verify that the application is indeed executing with the required hardware-derived protection. A set of trusted third parties participate to enable this attestation mechanism as in a standard Public Key Infrastructure mechanism. Our P-MAPS TCB is substantially smaller (~2500x smaller compared to a commodity OS) TCB. We continue to strive to reduce this TCB layer, and analyze requirements that different applications impose on the P-MAPS, as well as do performance analysis of the overheads while it executes.

Comments (0)