computation is increasingly moving to the data center. Thus, the energy used by CPUs in the data center is gaining importance. The centralization of computation in the data center has also led to much commonality betw...
详细信息
ISBN:
(纸本)9781479969982
computation is increasingly moving to the data center. Thus, the energy used by CPUs in the data center is gaining importance. The centralization of computation in the data center has also led to much commonality between the applications running there. For example, there are many instances of similar or identical versions of the Apache web server running in a large data center. Many of these applications, such as bulk image resizing or video transcoding, favor increasing throughput over single stream performance. In this work, we propose Execution Drafting, an architectural technique for executing identical instructions from different programs or threads on the same multithreaded core, such that they flow down the pipe consecutively, or draft. Drafting reduces switching and removes the need to fetch and decode drafted instructions, thereby saving energy. Drafting can also reduce the energy of the execution and commit stages of a pipeline when drafted instructions have similar operands, such as when loading constants. We demonstrate Execution Drafting saving energy when executing the same application with different data, as well as different programs operating on different data, as is the case for different versions of the same program. We evaluate hardware techniques to identify when to draft and analyze the hardware overheads of Execution Drafting implemented in an OpenSPARC T1 core. We show that Execution Drafting can result in substantial performance per energy gains (up to 20%) in a data center without decreasing throughput or dramatically increasing latency.
With the proliferation of connected devices including smartphones, novel network connectivity and management methods are needed to meet user Quality of Experience (QoE) and computational demands of contemporary applic...
详细信息
With the proliferation of connected devices including smartphones, novel network connectivity and management methods are needed to meet user Quality of Experience (QoE) and computational demands of contemporary applications. Ser vice caching and computation reuse techniques are being employed to alleviate challenges due to scalability, interoperability, and mobility, as well as to reduce application latency by enabling caching at the edge. This survey provides a taxonomy for service caching and computation reuse and describes the current state of the research and its challenges. This is the first survey that provides a comprehensive analysis and suggests future research directions on this topic.
暂无评论