SPECjvm2008 Performance Characterization

Shiv, K.; Chow, Kingsum; Wang, Yanping; Петроченко, Д. В.

doi:10.1007/978-3-540-93799-9_2

Cited by 49 publications

(33 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the case of statically-typed languages such as Java and Scala, the execution charateristics have become rather well understood, thanks to established metrics, benchmarks, and studies [9,11,[21][22][23]. In contrast, the execution characteristics of various dynamically-typed JVM languages have so far not been studied very extensively.…”

Section: Introductionmentioning

confidence: 99%

Characteristics of dynamic JVM languages

Sarimbekov

Podzimek

Bulej

et al. 2013

Proceedings of the 7th ACM Workshop on Virtual Machines and Intermediate Languages

View full text Add to dashboard Cite

The Java Virtual Machine (JVM) has become an execution platform targeted by many programming languages. However, unlike with Java, a statically-typed language, the performance of the JVM and its Just-In-Time (JIT) compiler with dynamically-typed languages lags behind purpose-built language-specific JIT compilers. In this paper, we aim to contribute to the understanding of the workloads imposed on the JVM by dynamic languages. We use various metrics to characterize the dynamic behavior of a variety of programs written in three dynamic languages (Clojure, Python, and Ruby) executing on the JVM. We identify the differences with respect to Java, and briefly discuss their implications.

show abstract

Section: Introductionmentioning

confidence: 99%

Characteristics of dynamic JVM languages

Sarimbekov

Podzimek

Bulej

et al. 2013

Proceedings of the 7th ACM Workshop on Virtual Machines and Intermediate Languages

View full text Add to dashboard Cite

show abstract

“…Jbb requires 6 GB/s on a server with four quad-core Intel Core-2 Duo processors [45]. In our environment, jbb requires about 10 GB/s.…”

Section: Workload Validationmentioning

confidence: 99%

Towards energy-proportional datacenter memory with mobile DRAM

Malladi

Lee

Nothaft

et al. 2012

SIGARCH Comput. Archit. News

View full text Add to dashboard Cite

To increase datacenter energy efficiency, we need memory systems that keep pace with processor efficiency gains. Currently, servers use DDR3 memory, which is designed for high bandwidth but not for energy proportionality. A system using 20% of the peak DDR3 bandwidth consumes 2.3× the energy per bit compared to the energy consumed by a system with fully utilized memory bandwidth. Nevertheless, many datacenter applications stress memory capacity and latency but not memory bandwidth. In response, we architect server memory systems using mobile DRAM devices, trading peak bandwidth for lower energy consumption per bit and more efficient idle modes. We demonstrate 3-5× lower memory power, better proportionality, and negligible performance penalties for datacenter workloads.

show abstract

“…They used SPECjvm2008 for evaluating the accuracy of their model. [17] analyzes the workload of SPECjvm2008 on high-end desktop processors (Core 2 Duo and Core i7) in various aspects, including Base vs Peak comparisons (i. e. without and with optimizations), clock per instruction and cache/TLB miss rates, memory allocation behavior and thread scaling. Seo et.…”

Section: Related Workmentioning

confidence: 99%

Power-performance analysis of JVM implementations

2011

ICIMU 2011 : Proceedings of the 5th International Conference on Information Technology &Amp; Multimedia

View full text Add to dashboard Cite

Abstract-Java Virtual Machines (JVMs) work between the Java application programs and the operating systems (and their underlying hardware platforms) to provide the 'writeonce run-anywhere' property of the Java language. However, this property also implies that the runtime efficiency, in terms of both performance and power-consumption, can be affected by the implementations of JVM.In this paper, we present a case study of performance-power analysis of JVM implementations. We run SPECjvm2008 with OpenJDK and IBM I9 on an Atom-based netbook with Ubuntu operating system. Our observations are as follows: (1) the relative performance of OpenJDK ranges from 44 to 289% of J9, (2) the dynamic power consumption ranges from 2.8 to 7.2 Watts among benchmark programs. However, the power consumptions of two JVMs for the same workload are relatively similar. Therefore, the power-performance efficiency is mostly affected by the relative performance. (3) the effectiveness of multi-threading varies among benchmark programs as well as among JVMs. In general, running benchmark with a single-thread loses more in performance than in power consumption. Exceptions are compress, fft.small (OpenJDK only) and lu.small. (4) for most benchmark programs (except scimark), the power consumptions seem to be correlated to linear and square root of L2 reference and L2 miss rates, respectively.

show abstract

SPECjvm2008 Performance Characterization

Cited by 49 publications

References 1 publication

Characteristics of dynamic JVM languages

Characteristics of dynamic JVM languages

Towards energy-proportional datacenter memory with mobile DRAM

Power-performance analysis of JVM implementations

Contact Info

Product

Resources

About