January 2, 2014
Adobe LiveCycle Engineering has the following update:
We have run some additional benchmark tests to compare the performance of LiveCycle on Oracle M-Series and T-Series servers, and found that the T5-4 server performs at par with the M3000 server in single- and multi-threaded scenarios (Output, Signatures).
This fact is also supported by the IBM’s Processor Value Unit(PVU) calculations. PVU is used by IBM to license its middleware and is based on the relative performance of processors. IBM assigns a PVU of 100 to T4-4, T5-4 and UltraSPARC IV. SPARC64 VI, VII and X processors also get a PVU of 100. SPARC M5 gets a PVU of 120 and the same is assigned to T5-8.
October 4, 2010
Urs Hölzle, Google’s Senior VP of Operations has published a very interesting article in the July/August 2010 “Datacenter Computing” issue of the journal IEEE Micro comparing the performance of what he calls “wimpy core” CPUs against “brawny core” CPUs. You can get it here.
According to this paper, a “wimpy core” CPU is one whose designers have compromised its single-thread performance in favour of massive parallelism and low power consumption. According to Urs, “…suppose a Web service runs with a latency of one second per user request, half of it caused by serial CPU time. If we switch to wimpy-core servers, whose single-threaded performance is three times slower, the response time doubles to two seconds and developers might have to spend a substantial amount of effort to optimize the code to get back to the one second latency…”
Currently, from a LiveCycle deployment perspective, the only “wimpy core” CPU servers we know of are Sun/Oracle’s T-series servers with the UltraSPARC T1 and T2 CPUs. We consider Sun/Oracle’s M-series servers with SPARC64 VII CPUs “brawny core”.