Multi-cache: Dynamic, Efficient Partitioning for Multi-tier Caches in Consolidated VM Environments

Abstract

Every physical machine in today's typical datacenter is backed by storage devices with hundreds of Gigabytes to Terabytes in size. Data center vendors usually use hard disk drives for their back-end storage as it is cheap and reliable. However, the increase in the I/O accesses to the back-end storage from one or many of the VMs hosted on a physical machine can reduce its overall accesses time significantly due to contention. This may not be suitable for interactive applications requiring low latency that might be co-located with other I/O intensive applications.In this paper we present Multi-Cache, a multi-layer cache management system that uses a combination of cache devices of varied speed and cost such as solid state drives, non-volatile memories, etc to mitigate this problem. Multi-Cache partitions each device dynamically at runtime according to the workload of each VM and its priority. We use a heuristic optimization technique that ensures maximum utilization of the caches resulting in a high hit rate.We use a weighted partitioning policy that improves latency by up to 72% for individual workloads, and a overall hit rate increase of up to 31% for host running several workloads together in comparison to standard LRU caching algorithms.

Proposal and Evaluation of IO Concentration-Aware Mechanisms to Improve Efficiency of Hybrid Storage Systems

“…First, we describe another piece of research in which cache capacities were dynamically assigned in accordance with changes in workload. Sundaresan et al [28] proposed Multi-Cache, a multi-layer cache management system that uses a combination of cache devices of varied speed and cost, such as SSDs and NVMs, to dynamically allocate cache capacities among different virtual machines (VMs). Multi-Cache partitions each device dynamically at runtime in accordance with the workload of each VM and its priority.…”

Cacol: A zero overhead and non-intrusive double caching mitigation system

“…Multi-Cache [21] is a multi-layer cache management system that uses a combination of different types of cache to ensure a minimum quality of service. In particular, it suggests an optimal combination of cache types for different speed disks (e.g.…”

Consideration for effectively handling parallel workloads on public cloud system

“…Cloud services are rapidly spreading because they can speed up software development for users, reduce a user's operational costs, and enable agile development. These services are built on virtual machine (VM) environments, and these VMs often compete for input/output (IO) performance [14]. It is important for us to improve the costperformance of cloud services by resolving the conflict of IO performance.…”