Evaluating Adaptive Compression to Mitigate the Effects of Shared I/O in Clouds

Abstract

IaaS clouds have become a promising platform for scalable distributed systems in recent years. However, while the virtualization techniques of such clouds are key to the cloud's elasticity, they also result in a reduced and less predictable I/O performance compared to traditional HPC setups. Besides the regular performance degradation of virtualized I/O itself, it is also the potential loss of I/O bandwidth through co-located virtual machines that imposes considerable obstacles for porting dataintensive applications to that platform.In this paper we examine adaptive compression schemes as a means to mitigate the negative effects of shared I/O in IaaS clouds. We discuss the decision models of existing schemes and analyze their applicability in virtualized environments. Based on an evaluation using XEN, KVM, and Amazon EC2, we found that most decision metrics (like CPU utilization and I/O bandwidth) are displayed inaccurately inside virtual machines and can lead to unreasonable levels of compression. As a remedy, we present a new adaptive compression scheme for virtualized environments which solely considers the application data rate. Without requiring any calibration or training phase our adaptive compression scheme can improve the I/O throughput of virtual machines significantly as shown through experimental evaluation.

Capacity planning of EFT service hosted on elastic IaaS

“…This activity also defines the set of metrics that will be estimated. Among such metrics, some might be highlighted such as processor and disk utilization, average disk response time, average disk queue length, available memory [7].…”

Monetary Cost Optimizations for Hosting Workflow-as-a-Service in IaaS Clouds

“…There have been some proposals to reduce the performance interference and unpredictability in the cloud, such as network performance [43] and I/O performance [44], [45], [46]. However, by the design of cloud computing, cloud is shared by many concurrent executions.…”

Collaborative Compressed I-cloud Medical Image Storage with Decompress Viewer

“…Since the concept is still new, no safety standard has actually been developed; each company is following its own standards. 16 Our proposed method includes some important security services such as identity based authentication, encryption, decryption together with compression and decompression in cloud computing system. We exploit the mechanism of encryption at the main cloud storage server using secret key followed by compression of images while uploading to an intermediate server.…”

“…and can be applied at a fixed set of n compression levels as a means to mitigate the effects of shared I/O to improve the efficiency of distributed applications which rely heavily on the network. 16 All the compression approaches mentioned above intend to conserve bandwidth and improve transfer speed, but are limited to end-to-end transfers rather than total aggregated throughput. Moreover, compression is applied in-transit only instead the data being stored remotely in a compressed fashion and therefore requests for the same data invoke new compression-decompression cycles every time.…”