After a short break, I'm back in blogland. In the meantime, I and my teamhave moved back from Sun's Cloud Computing Engineering organization toSun Studio (Compilers and Tools). It was a wonderful ride and I learnedso many things that I intend to build on, in coming months. Of course,my group is still involved in the same Cloud-related tools project ofmaking HW, SW stack and tools more easily accessible to developers whodont have OpenSolaris (or Solaris) on their desktop and may not haveaccess to a SPARC machine in their development group. More on that inthe coming weeks, but right now I'll turn to Sun Studio relatedactivities.

This was the week for IntelDeveloper Forum 09 (Sept 22-24, Moscone West, San Francisco).
Last year, the emphasis seemed to be on Nehalem, AVX, Graphics andParallelism.
This year, the emphasis seems to be around Mobility, some followup onParallelism and Cloud Computing. Intel is totally on top of the worldwith the Nehalem chip: a well-balanced, high performance chip with agreat feature set that the company can build their entire roadmap on.They are on a high, and know they have a winner in Nehalem.
This year again, wewere invited to have a booth and a Chalk Talk at the conference.The booth duty was interesting and you really get to do some deep-divetype conversations with some interesting folks who walk by and we gotour share, this year as well. Which makes it all worthwhile andstimulating. Its an ideal time to listen to what other developers haveto say about our products (both good and bad and we heard both sides)and to share views on where the environment is headed. If you remember,I gave a Chalk Talk last year as well. This year's talk wasin our own booth, so it was more lightly attended but it was fun (andchaotic) as well. My focus was on Compiler performance and the newWorld Records we have created since the launch of Nehalem systems (getdetails here: http://www.sun.com/benchmarks/software/index.jspand look for the Sun Studio logo), on new features (OpenMP 3.0, SSSE3,SSE4.1, SSE4.2, ), new parallelization assistance tools (DBXtool, MPIanalyzer, Profiling D-trace like with D-light and DTrace GUI), ease-of-development with a fully-integrated IDE (based on NetBeans 6.5with considerably enhanced C/C++ support) and continuous ongoingimprovements (lots of improvements on the performance side, with bettervectorizer, register allocator, instruction scheduler, etc, an improvedPerformance Analyzer and Thread Analyzer with support for new HWcounters and too many to describe here in details). Look up here for more details.

Intel itself build IDF as a showcase for next, next, next generation oftechnologies. What was truly interesting was how much focus there wason Cloud Computing. They had two dedicated 3-day track on this (one forPublic Cloud and another for Enterprise Cloud), but more than that, itwas interspersed at many of the other talks as well. The emphasis wasclearly on educating on technologies they provide to enhanceDatacenters:
  • Networking: with 10Gb controllers, reduced IO latency inNehalem and LAN/SAN interconnect fabric.
  • Storage: with a big push towards SSD/Flash (better iops/$) tocreate right balance between cost and IOps needs
  • Virtualization: HW assist, VM density, performance and powerefficiency, enhanced IO virtualization
  • Security enhancements by highlighting AES(Advanced Encryption)and TXT(Trusted Execution) features in the chips
  • Performance of Nehalem chips which allows for optimized SW andreduced datacenter costs
The impression I got was that they were pushing Clouds for Enterprisesthat needed hyper-scale efficiency that was utilitarian (rather thandifferentiating) with homogeneous HW with greater focus on cost ofinitial ownership (rather than TCO, they arent convinced that Cloudsdifferentiate on TCO). In fact, IMO, their view was stronglydatacenter-centric, rather than Cloud as an elastic, available,multi-tenant, heterogenous, business-critical, differentiating sort ofview. Not that they didnt think these issues werent important, but itlooked like they werent going to address them as they werent core toIntel. Fair enough, but its important to know how one of the primarytechnology providers view this.



More...