Recently, flash-based solid-state drives (SSDs) have become standard options for laptop and desktop storage, but their impact on enterprise server storage has not been studied. Provisioning server storage is challenging. It requires optimizing for the performance, capacity, power and reliability needs of the expected workload, all while minimizing financial costs. In this paper we analyze a number of workload traces from servers in both large and small data centers, to decide whether and how SSDs should be used to support each. We analyze both complete replacement of disks by SSDs, as well as use of SSDs as an intermediate tier between disks and DRAM. We describe an automated tool that, given device models and a block-level trace of a workload, determines the least-cost storage configuration that will support

Bursts in data center workloads are a real problem for storage subsystems. Data volumes can experience peak I/O request rates that are over an order of magnitude higher than average load. This requires significant overprovisioning, and often still results in significant I/O request latency during peaks. In order to address this problem we propose Everest, which allows data written to an overloaded volume to be temporarily off-loaded into a short-term virtual store. Everest creates the short-term store by opportunistically pooling underutilized storage resources either on a server or across servers within the data center. Writes are temporarily off-loaded from overloaded volumes to lightly loaded volumes, thereby reducing the I/O load on the former. Everest is transparent to and usable by unmodified applications, and does not change the persistence or consistency of the storage system. We evaluate Everest using traces from a production Exchange mail server as well as other benchmarks: our results show a 1.4–70 times reduction in mean response times during peaks. 1

Multi-tier systems that combine SSDs with SAS/FC and/or SATA disks mitigate the capital cost burden of SSDs, while benefiting from their superior I/O performance per unit cost and low power. Though commercial SSD-based multi-tier solutions are available, configuring such a system with the optimal number of devices per tier to achieve performance goals at minimum cost remains a challenge. Furthermore, these solutions do not leverage the opportunity to dynamically consolidate load and reduce power/operating cost. Our extent-based dynamic tiering solution, EDT, addresses these limitations via two key components of its design. A Configuration Adviser EDT-CA determines the adequate mix of storage devices to buy and install to satisfy a given workload at minimum cost, and a Dynamic Tier Manager EDT-DTM performs dynamic extent placement once the system is running to satisfy performance requirements while minimizing dynamic power consumption. Key to the cost minimization of EDT-CA is its ability to simulate the dynamic extent placement afforded by EDT-DTM. Key to the overall effectiveness of EDT-DTM is its ability to consolidate load within tiers when feasible, rapidly respond to unexpected changes in the workload, and carefully control the overhead due to extent migration. Our results using production workloads show that EDT incurs lower capital and operating cost, consumes less power, and delivers similar or better performance relative to SAS-only storage systems as well as other simpler approaches to extent-based tiering. 1

Elasticity of cloud computing environments provides an economic incentive for automatic resource allocation of stateful systems running in the cloud. However, these systems have to meet strict performance Service-Level Objectives (SLOs) expressed using upper percentiles of request latency, such as the 99th. Such latency measurements are very noisy, which complicates the design of the dynamic resource allocation. We design and evaluate the SCADS Director, a control framework that reconfigures the storage system on-the-fly in response to workload changes using a performance model of the system. We demonstrate that such a framework can respond to both unexpected data hotspots and diurnal workload patterns without violating strict performance SLOs. 1

Good execution of data placement, caching and consistency policies across a user’s personal devices has always been hard. Unpredictable networks, capricious user behavior with leaving devices on or off and non-uniform energy-saving policies constantly interfere with the good intentions of a storage system’s policies. This paper’s contribution is to better manage these inherent uncertainties. We do so primarily by building a low-power communication channel that is available even when a device is off. This channel is mainly made possible by a novel network interface card that is carefully placed under the control of storage system protocols. The design space can benefit existing placement policies (e.g., Cimbiosys [21], Perspective [23], Anzere [22]). It also allows for interesting new ones. We build a file system called ZZFS around a particular set of policies motivated by user studies. Its policies cater to users who interact with the file system in an ad hoc way — spontaneously and without pre-planning. 1

We develop a holistic framework for adaptively scheduling asynchronous requests in distributed file systems. The system is holistic in that it manages all resources, including network bandwidth, server I/O, server CPU, and client and server memory utilization. It accelerates, defers, or cancels asynchronous requests in order to improve application-perceived performance directly. We employ congestion pricing via online auctions to coordinate the use of system resources by the file system clients so that they can detect shortages and adapt their resource usage. We implement our modifications in the Congestion-Aware Network File System (CA-NFS), an extension to the ubiquitous network file system (NFS). Our experimental result shows that CA-NFS results in a 20 % improvement in execution times when compared with NFS for a variety of workloads. 1

Abstract. The rapid increase in the amount of available information from various online sources poses new challenges for programs that endeavor to process these sources automatically and identify the most relevant material for a given application. This paper introduces an approach for optimizing queries to Semantic Web resources based on ideas originally proposed by MacQueen for optimal stopping in business economics. Modeling applications as decision makers looking for optimal action/answer sets, facing search costs for acquiring information, test costs for checking these information, and receiving a reward depending on the usefulness of the proposed solution, yields strategies for optimizing queries to external services. An extensive evaluation compares these strategies to a conventional coverage based approach, based on real world response times taken from popular Web services. 1