دانلود رایگان مقاله مدل های حجم کار و ارزیابی عملکرد خدمات ذخیره سازی ابر

Abstract

Cloud storage systems are currently very popular with many companies offering services, including worldwide providers such as Dropbox, Microsoft and Google. These companies as well as providers entering the market could greatly benefit from a deep understanding of typical workload patterns their services have to face in order to develop cost-effective solutions. Yet, despite recent studies of usage and performance of these systems, the underlying processes that generate workload for the system have not been deeply studied. This paper presents a thorough investigation of the workload generated by Dropbox customers. We propose a hierarchical model that captures user sessions, file system modifications and content sharing patterns. We parameterize our model using passive measurements gathered from fourdifferent networks. Next, we use the proposed model to drive the development of CloudGen, a new synthetic workload generator that allows the simulation of the network traffic created by cloud storage services in various realistic scenarios. We validate CloudGen by comparing synthetic traces with actual data from operational networks. We then show its applicability by investigating the impact of the continuing growth in cloud storage popularity on bandwidth consumption. Our results indicate that a hypothetical 4-fold increase in both user population and content sharing could lead to 30 times more network traffic. CloudGen is a valuable tool for administrators and developers interested in engineering and deploying cloud storage services.

8. Conclusion and future work

This paper investigated the workload of cloud storage services. We proposed a model to describe the Dropbox workloads, and parameterized it using data from four different networks. We then used our model to implement CloudGen, the first workload generator that simulates both users’ behavior and the network traffic of cloud storage. We validated our workload generator, and illustrated its applicability by exploring the traffic volume that can be expected in future usage scenarios. We offer CloudGen to the community as free software, as a valuable tool to simulate the behavior of cloud storage clients. The analyses of Dropbox workloads from real data showed that our model for the Dropbox functioning is robust. Indeed, we found close agreement for all components of the model in four distinct datasets. We envision several directions for future work. Regarding our workload generator, we plan to extend it to other cloud storage protocols and terminals. For instance, we plan to add components to control the level of replication of updates, modeling cases where only part of peer-devices receives them. We also intend to model seasonality to simulate long-term dynamics of workloads in cloud storage as well as to plug CloudGen in well-known network simulators, then testing the impacts of network conditions on workloads. Moreover, we will investigate how individual user’s profile in- fluences the workload, and factor that into CloudGen, to allow the simulation of populations with particular characteristics. Finally, regarding applications of our model, we will leverage CloudGen to evaluate trade-offs (e.g., return on investments) of alternative architectures for cloud storage services.