LBICA: A Load Balancer for I/O Cache Architectures

Abstract

In recent years, enterprise Solid-State Drives (SSDs) are used in the caching layer of high-performance servers to close the growing performance gap between processing units and storage subsystem. SSD-based I/O caching is typically not effective in workloads with burst accesses in which the caching layer itself becomes the performance bottleneck because of the large number of accesses. Existing I/O cache architectures mainly focus on maximizing the cache hit ratio while they neglect the average queue time of accesses. Previous studies suggested bypassing the cache when burst accesses are identified. These schemes, however, are not applicable to a general cache configuration and also result in significant performance degradation on burst accesses.In this paper, we propose a novel I/O cache load balancing scheme (LBICA) with adaptive write policy management to prevent the I/O cache from becoming performance bottleneck in burst accesses. Our proposal, unlike previous schemes, which disable the I/O cache or bypass the requests into the disk subsystem in burst accesses, selectively reduces the number of waiting accesses in the SSD queue and balances the load between the I/O cache and the disk subsystem while providing the maximum performance. The proposed scheme characterizes the workload based on the type of in-queue requests and assigns an effective cache write policy. We aim to bypass the accesses which 1) are served faster by the disk subsystem or 2) cannot be merged with other accesses in the I/O cache queue. Doing so, the selected requests are responded by the disk layer, preventing from overloading the I/O cache. Our evaluations on a physical system shows that LBICA reduces the load on the I/O cache by 48% and improves the performance of burst workloads by 30% compared to the latest state-of-the-art load balancing scheme.

A New Readahead Framework for SSD-based Caching Storage in IoT Systems

“…Thus, current SSD cache managers adopt a selective caching policy, in which only blocks that are identified as valuable enough are cached into the SSD. That is, they treat the SSD as a by-passable cache to minimize unnecessary writes to the SSD [11][12][13]. Since HDDs exhibit poor performance with random accesses, many SSD cache managers differentiate randomly accessed data from sequentially accessed data and only cache randomly accessed blocks in the SSD.…”

LQoCo: Learning to Optimize Cache Capacity Overloading in Storage Systems

“…In Figure 1, a storage system typically consists of host-side servers, caching-tier and storage tier [6,7]. In general, host-side is comprised of general-purpose servers, which issue I/O requests to the cachingtier.…”

Evaluating Reliability of SSD-Based I/O Caches in Enterprise Storage Systems

“…SWM generates various workloads with defined parameters such as request size, request type, access pattern, sequence of accesses, and WSS and issues the I/O requests to the disk subsystem (i.e., HDD equipped with SSD-based I/O cache configuration). 6 SWM tracks the issued I/O requests and detects the parameters such as issue time, completion time, request size, request type, and the checksum of written data on disk. In addition, SWM keeps the 6.…”

“…Increasing the I/O intensive applications such as Online Transaction Processing (OLTP) and banking services makes storage subsystems built upon Hard Disk Drives (HDDs) as the performance bottleneck of enterprise systems. In order to alleviate the performance shortcomings of HDD-based storage subsystems, enterprise manufacturers such as Dell EMC, NetApp, and HP [1][2][3] and emerging storage architectures [4][5][6] employ high performance flash-based devices such as enterprise Solid-State Drives (SSDs) as a cache layer for disk subsystem, which is mainly composed from low-performance HDDs and mid-range flash-based SSDs (as depicted in Fig. 1).…”