Over the last decade, research in Electronic System Level (ESL) design has resulted in significant advances in addressing the rising design complexity and meeting the required performance constraints. Now a major concern of system-level design is the power reduction and energy dissipation of the system-on-chip which not only affects battery lifetime but also thermal aspects and reliability of t...

As energy has become one of the key operating costs in running a data center and power waste commonly exists, it is essential to reduce energy inefficiency inside data centers. In this paper, we develop an innovative framework, called PowerTracer, for diagnosing energy inefficiency and saving power. Inside the framework, we first present a resource tracing method based on request tracing in mul...

Virtualized infrastructure providers demand new methods to increase the accuracy of the accounting models used to charge their customers. Future data centers will be composed of many-core systems that will host a large number of virtual machines (VMs) each. While resource utilization accounting can be achieved with existing system tools, energy accounting is a complex task when per-VM granulari...

This paper discusses the benefit of dynamic voltage and frequency scaling for scientific applications under different optimization levels. The reported experiments show the tradeoffs between optimizations for performance in terms of execution time, power dissipation, and overall energy consumption.

In this paper, a runtime performance projection model for dynamic power management is proposed. The model is built as a first-order linear equation using a linear regression model. It could be used to estimate performance impact from different p-states (voltage-frequency pairs). Workload behavior is monitored dynamically for a program region of 100M instructions using hardware performance monit...

This paper presents a method to reduce the energy of interactive systems for mixed workloads: multimedia applications that require constant output rates and sporadic jobs that need prompt responses. The authors’ method divides multimedia programs into stages and inserts data buffers between them. Data buffering has three purposes: 1) to support constant output rates; 2) to allow frequency scali...

/ 0214365 7 5&8 9;:28 3 < 9 1>= 8 3?7A@ BC9 DE7%FG@ 8 8-:H= 7H< I J < B K&LM9-B%FN7 F OHB 9 1PLP0 DQ@ B 9 :CR K 7 DS1><#TCB F LU7 O28 V @ 9-8 W X 8 K < R 3 < 7%F 1PK OZY4[Z\ ]%^;14K_7`5 B 9 LU7H= F 8#DSX F J LM1PDa8-: 1>7b3 RC3 Lc8 DdLUBb3 7 T 8E5 B ef8 9geh0 1MF 8_D 7C14K LU7 1PK 14K2Oi7j:28 3 149 8-: 5 F 7 RH= 7H< IE9-7 Lc8 kmlfX 9n7%F OHB 9 14LP0 Do143`3 1PDS1MF 7 9`14Kp< B D`5qF 8-r 14LMR L...

Workload consolidation is usually performed in datacenters to improve server utilization for higher energy efficiency. One of the key issues in workload consolidation is the contention for shared resources. Dynamic voltage and frequency scaling (DVFS) of CPU is another effective technique that has been widely used to trade performance for power reduction. We have found that the degree of resour...

The module-wise dynamic voltage and frequency scaling (MDVFS) scheme is applied to a single-chip H.264/MPEG-4 audio/visual codec LSI. The power consumption of the target module with controlled supply voltage and frequency is reduced by 40% in comparison with the operation without voltage or frequency scaling. The consumed power of the chip is 63 mW in decoding QVGA H.264 video at 15 fps and MPE...

Recurrent neural networks (RNNs) are widely used to process sequence-related tasks such as natural language processing. Edge cloud computing systems in an asymmetric structure, where task managers allocate the edge and based on computation requirements. In a system, servers have no energy limitations, since they unlimited resources. systems, however, resource-constrained, consumption is thus ex...