Kiran Srinivasan

Design implications for enterprise storage systems via multi-dimensional trace analysis

Enterprise storage systems are facing enormous challenges due to increasing growth and heterogeneity of the data stored. Designing future storage systems requires comprehensive insights that existing trace analysis methods are ill-equipped to supply. In this paper, we seek to provide such insights by using a new methodology that leverages an objective, multi-dimensional statistical technique to extract data access patterns from network storage system traces. We apply our method on two large-scale real-world production network storage system traces to obtain comprehensive access patterns and design insights at user, application, file, and directory levels. We derive simple, easily implementable, threshold-based design optimizations that enable efficient data placement and capacity optimization strategies for servers, consolidation policies for clients, and improved caching performance for both.

An empirical study of file systems on NVM

Emerging byte-addressable, non-volatile memory like phase-change memory, STT-MRAM, etc. brings persistence at latencies within an order of magnitude of DRAM, thereby motivating their inclusion on the memory bus. According to some recent work on NVM, traditional file systems are ineffective and sub-optimal in accessing data from this low latency media. However, there exists no systematic performance study across different file systems and their various configurations validating this point. In this work, we evaluate the performance of various legacy Linux file systems under various real world workloads on non-volatile memory (NVM) simulated using ramdisk and compare it against NVM optimized file system - PMFS. Our results show that while the default file system configurations are mostly sub-optimal for NVM, these legacy file systems can be tuned using mount and format options to achieve performance that is comparable to NVM-aware file system such as PMFS. Our experiments show that the performance difference between PMFS and ext2/ext3 with execute-in-place (XIP) option is around 5% for many workloads (TPCC and YCSB). Furthermore, based on the learning from our performance study, we present few key file system features such as in-place update layout with XIP, and parallel metadata and data allocations, etc. that could be leveraged by file system designers to improve performance of both legacy and new file systems for NVM.

Responding rapidly to service level violations using virtual appliances

One of the key goals in the data center today is providing storage services with service-level objectives (SLOs) for performance metrics such as latency and throughput. Meeting such SLOs is challenging due to the dynamism observed in these environments. In this position paper, we propose dynamic instantiation of virtual appliances, that is, virtual machines with storage functionality, as a mechanism to meet storage SLOs efficiently. In order for dynamic instantiation to be realistic for rapidlychanging environments, it should be automated. Therefore, an important goal of this paper is to show that such automation is feasible. We do so through a caching case study. Specifically, we build the automation framework for dynamically instantiating virtual caching appliances. This framework identifies sets of interfering workloads that can benefit from caching, determines the cache-size requirements of workloads, non-disruptively migrates the application to use the cache, and warms the cache to quickly return to acceptable service levels. We show through an experiment that this approach addresses SLO violations while using resources efficiently.