A few months ago Mark Polly and I did a sizing for Websphere Portal on IBM Power 7 processor. In less than 24 hours IBM’s Techline came back with a sizing that was about 60% of what we were expecting to see. In other words, using Power 7 processors caused 40% reduction in the number of processors we were expecting. That’s vs a Power 6 architecture. It blew us away. We actually went back to Techline and asked them if they made a mistake. They said, “Nope, that’s the correct sizing for Power 7”
Now IBM uses a very sophisticated model to figure out what to charge based on the various processor architectures and types out there. Even with a slight uptick in cost per processor core, you still get a huge amount of extra value by going with Power 7’s vs any of the older AIX machines. It’s such a big difference that it’s scary.
I did some research on why the difference is so huge and came across an article at the Register reviewing what’s new and why Power 7 is so much faster.
There are a couple of big changes with the Power7 design, and all of them impact performance. First and foremost, the chip includes 32 MB of on-chip L3 cache memory implemented in embedded DRAM instead of the off-chip L3 cache that was used with all the prior dual-core Power chips. This, as it turns out, may be more important than boosting the threads and cores compared to the Power6 and Power6+ chips.
………The effect of this eDRAM on the Power7 design, and its performance, is two-fold. First, by adding the L3 cache onto the chip, the latency between the cores and the L3 memory has been reduced by a factor of six, according to Handy. (The exact memory latency feeds and speeds were not available at press time). This means the Power7 cores are waiting a lot less for data than the previous Power cores were. Also, by having that L3 cache take up a lot less space than it might otherwise, IBM could boost the core count by a factor of four, to eight cores on a die, and could double the thread count per core, to four. If it were not for the eDRAM, the Power7 chip might have looked a lot like Tukwila, with its transistor budget being half burned up by cache.
…….Equally importantly for an IBM that is doing battle with Oracle and its Sparc T 64-threaded T2 and T2+ chips and the quad-core, eight-threaded Tukwilas due from Intel today, the Power7 chip has 32 threads, eight times as many as the Power5 through Power6+ chips could bring to bear on workloads that like threads. One of those workloads is IBM’s own WebSphere Application Server, and on early benchmark tests, shifting from a Power6 to Power7 system with the same number of cores boosted the performance of WebSphere running on AIX by 85 per cent.
It’s worth it to read the entire article but it blows me away that you can get up to an 85% boost running on WebSphere Application Server (WAS). Since WebSphere Portal runs on WAS, I can see why the sizing is so different.
Now all that said, we still have seen Portal running in Production yet and haven’t yet completed real life load testing. When we do, I’m sure Mark or I will blog about it.
