Neelesh is my childhood friend. We recently got connected after some 20 years – thanks to Web2.0… To my last post “High Perfromance Computing – not the same anymore”, he made a very interesting comment and it cannot stay tucked inside the comment corner of my blog but be up here for everyone to read. Neelesh is now based in NY and works for Citibank as Vice President (I can now guess VP for what?).
Thanks Neelesh!
====================================
Uttam,
Its interesting you mention about CUDA and the advent of GPU based computing.
This is something most of the high-end investment banking houses here in NYC and also in the UK are seriously looking at. We currently use symphony based grid orchestration for lot of our compute power needs for various risk calculations (monte-carlo simulations, prepayment modeling and fixed income risk calculations).
But now several teams are trying to evaluate moving to GPU based computing. My initial reactions to this was that GPU based computing is good if you are doing a “pure-maathematical-grind” type of calculations rather than something that requires an “if-else” type structured logic. Well that does not come to me as a surprise as GPUs were primarily built for that kind of stuff – image rendering or graphics type programming, where matrix algebra is the cornerstone for all computations.
What we were a little hesitant as part of our evaluation and future possible adoption was that someone using CUDA is sort of married to the architecture and its hard to unplug out of it. Most of the code logic in CUDA world is very intrusive and not platform agnostic. Might not be a big issue but this is something software engineers are usually cautious about. (P.S: we were looking at the NVIDIA based products).
But on the flip side there is huge-huge cost savings in hardware needs and data center needs. We were amazed by the amount of reduction we would have in h/w if we ever redid all of our code using the CUDA based API.
While this space is sure to evolve in the very near future in terms of compute power needs, I also have a strong suspicion that the likes of Intel/AMD will try to put something in place in their architectures to prevent or limit folks from leveraging the massive GPU power or at least make sure that they are somehow the part of the equation. At the end of the day the “instruction-set” or the brain power still is with the main CPU and they (Intel/AMD) could tweak things around…
Scientific and Enterprise computing that are CPU intensive is not going to be same anymore. GPU architectures are making those large G/T/P FLOP systems faster, greener and smaller, also cheaper.
More and more applications are getting accelerated with CUDA. Among good examples are – CFD solvers running at 1/20 of resource for a given iteration; HMMER running at 62x time faster; Monte Carlo Simulations running up to 50x time faster.
The GPU architectures from CUDA and FireStream have already made it to Top500 list sometime ago which has been a milestone and is fast moving up the charts, I am sure the 35th list of Top500 is going have much more of GPU+CPU. It is only matter of time that desktops would be replaced with supercomputers. The GPU’s help offloading non-sequential parts of application from the CPU and achieves performance of nature described before.
More interesting to watch is if CUDA is really breaking the dominance of Intel and AMD? And if more and more developers will optimize applications with OpenCL or DirectX for CUDA? What will even happen to ClearSpeed?
Here too my paper got selected and am speaking at the Bangalore event.
The topic is same as the one I spoke at INTEROP but this time around, I will touch base on “Service Orientation for the Enterprise Cloud”. You may wonder, cloud is all about service orientation anyway so what would SOC (my musings on SOx, after SOI & SOM) – Service Oriented Cloud – be about.
As cloud definitions and perspectives evolve, we found our way to explain what goes within and beyond.
We are organising an event on 20-Aug-2009. The event will cover Security framework in the service-oriented economy.
You will hear from us about the “critical shift” with IdM and Access Management in today’s environment that is brought on by increased service standards, improved systems and business agility.
If SOA is the way to go, understanding the “critical shift” may be equally important.
Since I could not create one for our way of SOI/SOM, I borrowed this one and is the closest we are to. This is something I otherwise keep saying with product neutralism; however would never mind using Cisco technology in our stack to SOI. Go thru by pressing the replay button…
Next week Cisco would announce their Unified Computing strategy codenamed “California”. These are blade servers that would sit right in middle of the Data Centre unlike all other Cisco products that found place in the network closet.
This in good sense is SOI we wanted to achieve and were seeking product maturity for levels of automation it would demand.
In our definition of SOI Orchestration is most vital and hopefully would soon be available across all Data Centre components. I am upbeat about the Cisco announcement.
It is matter of immense responsibility and pride for all of us at Locuz. Sanjeev should be credited with the fact that his decision then helps Samit stay healthy thru tough times like these.
Every single meeting I had in the last two months – that I did not post anything on my blog – was about IT heads saying they have their eyes fixed on the ball – doing more or same but certainly with less.
Remote Infrastructure Management is the surest way. But this can be misleading too. It is extremely important to know what capabilities one should look for in the RIM partner. Let me offer a few tips (not in any order and none being optional) -
1. RIM partner should be able to offer flexible models
2. Should have deep Network and Systems Integration skills
3. Be specialized in Security
4. Contribute in transformational projects for Business IT alignment
5. Offer best practices and have delivery Centre Certified
6. Introduce Automation and Standardization – processes and delivery
7. Should be able to co-locate systems closest to the Delivery team
8. Virtual staff augmentation of highest quality
9. Contribute to Architecture decisions
10. Innovate (and can mean so much)
11. Tools – best of breed
12. Improving Datacentre efficiency for power and space
13. Consolidate and optimize the network, systems and storage
14. Mitigate all risks connected with Business Services availability
15. Finally bring cost down by 20% in the first year
Our customers have chosen us for these reasons.
Conclusion – RIM Services is the way forward to better manage IT Infrastructure while reducing operational costs.
I will have a note from a customer who is keen to share the experience on my next post.
But then are taking the responsibility of innovating with our customers who will eventually brave this crisis. At times like these organisations are seeking partnerships that will cut cost right from the day of engagement and not thru projects that will give returns over the medium to long term.
Our business models and methodologies are very compelling and made with customers in mind for good times and to stay smiling at bad times.
You will not have to stall a critical IT project but speak to us; we will deliver all of that to you as service and feel free to call it SaaS, HaaS or Cloud.
Our delivery for Selective Outsourcing will be Integratable to your existing Enterprise systems because of our belief in Adaptive Enterprise and unique SOI/SOM methodology. We will deploy your services over an orchestration layer and manage it thru SLA driven high availability, security and performance from our IDC at Chennai which is among the best in India.
OR
A model that delivers SLA performance to manage your existing environment leveraging our RIM (Remote Infrastructure Management) capabilities combined with superior levels of technology, processes and service delivery automation. You may spend 20-30% lower.
Spend less and run Business- IT dashboard (see to believe) with improved analysis of root cause, event correlation and problem management.
OR
Give away your DR problems to us. We are very special when it comes to offering you Managed DR.
We are spending lot of time in thinking and piloting on how else we could innovate. IdM/SSO is available as SaaS with us now; more of such innovation is going to be soon rolling out. May be we could co-create for your environment. Let us talk.
Quite catchy - Cisco way to Unified Fabric Architecture… Its promise was to deliver flexibility across Physical, Virtual and Cloud environments for any application – now promises enhancements for new age massively scalable Data Centres to deal with Big Data/Hadoop environments and bare-metal deployments. These data centres have massive blade density and dema […]
The rapidly growing use of Tablets by Enterprise users is making it imperative they get anywhere any device access to all applications. Anytime is implicit. Many Businesses are unprepared with such sudden rise in on-demand access by unconventional devices. Access to Business applications had been in some form tied down to desktops/laptop hardware and operati […]
We are an authority in building Private Cloud Infra. Something that I have been longing to write for a while but wanted to be sure if we were not missing anything. With a tested methodology and deep expertise across parts of a service ready stack, in place, we can help customers transition their Core Technology […]
Locuz has won few CRN awards during the previous years, all being in the cat of “Best SI, South India”. Today we have been conferred with “COMPANY OF THE YEAR” Award by CRN – a very special National and highest echelon Award. It reflects our commitment to Quality Services, Technology Vision, HR Practices, Unique Skills, […]
This Morning I decided will blog about 8 points that to my mind seem some kind of predictions and just then saw a Google Alert message that had 14 predictions from Chirag Mehta and Wang. They said more than what I was sure and could not have agreed more – then just copeid them over […]
Not one CIO or CTO is avoiding the thought of cloud adoption – plain truth. There could be some who do not have a Business-IT posture to adopt cloud early but nevertheless are sitting on a transformation blueprint that is made with cloud and just cloud in the mind. Among the thought leaders who have […]
Teaching and Learning Engineering is not going to be same anymore in India. Some initiatives are redefining them and among these NPTEL is one. We are proud to be associated with VTU in Karnataka assisting them in establishing their on-demand video lecture provisioning system – connected with NPTEL and in-house studio. We used a uniquely […]
Identity Management is the corner stone of Governance and more so in the case of SOA. SOA interactions impose tremendous complexity on being able to make each interaction accountable from user access standpoint and unlike Identity in monolith applications; Identity for each interaction will have to be brokered effectively. It may be useful to plan […]
Recent Comments