Performance-directed energy management using BOS

Abstract

One of the major challenges in today's computing world is energy management in portable devices and servers. Power management is essential to increase battery life. High end server systems use large clusters of machines that consume enormous amount of power. Past research has devised both software and hardware techniques to memory energy management but has overlooked the performance of applications in such environments. The result is that some of these techniques slowed down an application by 835%. In this paper, we look at software techniques for memory energy management without compromising on performance. The paper conceives of a new approach called BOS - Ballooning in the OS inspired from the VMware ESX server. The BOS approach consists of a kernel daemon which continuously monitors the accesses to memory chips and disk I/O. Based on the profiled information, the BOS daemon decides about powering down/up chips. Powering down is emulated within the kernel using mechanisms such as page migration and invisible buddy. Results indicate that chips with more allocated pages may not always be the most frequently accessed ones. A study has been done analyzing the effect of decreased memory size on disk activity and based on the study, a threshold based policy is proposed which is found to settle in the operating point for a simple applicaton. A single page migration incurs a cost of approximately 13μs and is one of the bottlenecks in the BOS approach.

Energy Efficient Buffer Cache Replacement for Data Servers

“…Through judiciously selecting victim blocks, all memory accesses are naturally clustered into a small set of memory chips. Our algorithms differ from existing studies [2], [3], [7], [8], [9], [10], [11] in that ours do not rely on data migration and also have much less bookmarking overheads.…”

“…Ref. [2], [3], [7], [8] proposes to dynamically migrate hot data blocks to a smaller set of memory chips. Typically these approaches unavoidably pay fairly large performance and energy overhead during both the bookmarking process for identifying hot blocks and the migration course.…”

Energy and thermal aware buffer cache replacement algorithm

“…Instead of evicting out the block that will be accessed potentially in the farthest future, our algorithm evicts the one that are suboptimal in terms of hit rates but potentially has a larger energy saving. Our algorithm differs from existing studies [2], [4], [8], [9], [10], [11], [12] in that ours does not rely on data migration and thus has much less bookmarking and running overheads.…”

Energy-efficient memory management in virtual machine environments

“…Delaluz et al proposed an automatic data migration strategy that dynamically places the arrays with temporal affinity into the same set of banks and allows the use of more aggressive energy-saving modes [11]. Ramamurthy et al utilized page migration mechanism in performance-directed energy management [12]. Page migration can be very expensive due to time and energy overhead to move data and also the overheads by tracking mechanisms to classify page utilization.…”