Despite recent advancements and improved parallelism in multi-core CPU performance, there is still a big challenge to be solved relating to the scale-out of cloud applications.
Put simply, Linux application performance scales poorly as CPU core count increases. This is commonly experienced as typical Linux applications can be expected to see a 1.5X performance improvement for a 2-core CPU, but the scale quickly plateaus after that, with 4 core performance only improving around 2.5X. The performance further degrades as core counts rise. Given that, along with Intel’s announcement that its Xeon chips have up to 22 cores, scaling performance efficiently across cores is extremely important.
