Commercial workload performance in the IBM POWER2 RISC System/6000 processor

Abstract

We describe features of the POWERS" processor and memory subsystem that enhance RISC System/6000@' performance on commercial workloads. We explain the performance characteristics of commercial workloads and some of the common benchmarks used to measure them. Our own analysis methods are also described. Characteristics of commercial workloads Commercial workloads include a wide variety of applications; some of the more prominent include on-line transaction processing, other database management Wopyright 1994 by International Business Machines Corporation. Copying in printed form for private use is permitted without payment of royalty provided that (1) each reproduction is done without alteration and (2) the Journal reference and IBM copyright notice are included on the first page. The title and abstract, but no other portions, of this paper may be copied or distributed royalty free without further permission by computer-based and other information-service systems. Permission to republish any other portion of this paper must be obtained from the Editor.

An analysis of database workload performance on simultaneous multithreaded processors

“…Overall, this data confirms the aggregate cache behavior of transaction processing workloads found by others; namely, that they suffer from higher miss rates than scientific codes (at least as exhibited by SPEC and SPLASH benchmarks), with instruction misses a particular problem [3,6,12,16]. For example, columns 2 and 3 of Table 3 show that on-chip caches are relatively ineffective both at current cache sizes (32KB) and at larger sizes (128KB) expected in next-generation processors.…”

“…Franklin, et al, [6] identified the scarcity of loops and context switches as contributors to high instruction cache miss rates in commercial applications. Maynard, et al, [12] highlighted the large instruction footprints and high instruction cache miss rates of OLTP workloads.…”

“…Despite their increasing prominence, however, database management systems (DBMS) have been the subject of only limited architectural study [3,6,12,16,22]. Not surprisingly, these studies have shown that database systems can exhibit strikingly high cache miss rates.…”

An analysis of database workload performance on simultaneous multithreaded processors

“…While some previous studies have found that operating system kernel activity can dominate execution time for OLTP workloads [6,12,16], we found that a well-tuned workload spends most of its time in user-level code. Using DCPI, we determined that for OLTP, roughly 70% of execution time was spent in user-level code, with the rest in the kernel and the idle loop.…”

An architectural evaluation of Java TPC-W

“…These benchmarks have been very useful for gauging the absolute performance and price/performance of combined software/hardware systems. A significant body of prior work has studied the architectural requirements of such workloads [13,15,17,19,20,22]. At the same time, the Systems Performance Evaluation Cooperative (SPEC), a similar consortium, has developed standard benchmarks for evaluating both static and dynamic web content serving (SPECweb96 and SPECweb99, respectively) [26].…”