Avoiding the Streetlight Effect: I/O Workload Analysis with SSDs in Mind

Storage systems are designed and optimized relying on wisdom derived from analysis studies of file-system and block-level workloads. However, while SSDs are becoming a dominant building block in many storage systems, their design continues to build on knowledge derived from analysis targeted at hard disk optimization. Though still valuable, it does not cover important aspects relevant for SSD performance. In a sense, we are "searching under the streetlight", possibly missing important opportunities for optimizing storage system design. We present the first I/O workload analysis designed with SSDs in mind. We characterize traces from four repositories and examine their 'temperature' ranges, sensitivity to page size, and 'logical locality'. Our initial results reveal nontrivial aspects that can significantly influence the design and performance of SSD-based systems.

[1]  Eitan Yaakobi,et al.  Write Once, Get 50% Free: Saving SSD Erase Costs Using WOM Codes , 2015, FAST.

[2]  Antony I. T. Rowstron,et al.  Write off-loading: Practical power management for enterprise storage , 2008, TOS.

[3]  Raju Rangaswami,et al.  I/O Deduplication: Utilizing content similarity to improve I/O performance , 2010, TOS.

[4]  Mahesh Balakrishnan,et al.  Extending SSD Lifetimes with Disk-Based Write Caches , 2010, FAST.

[5]  Thomas E. Anderson,et al.  A Comparison of File System Workloads , 2000, USENIX Annual Technical Conference, General Track.

[6]  Peter Desnoyers,et al.  Analytic Models of SSD Write Performance , 2014, TOS.

[7]  Shankar Pasupathy,et al.  Measurement and Analysis of Large-Scale Network File System Workloads , 2008, USENIX Annual Technical Conference.

[8]  Eitan Yaakobi,et al.  It's Not Where Your Data Is, It's How It Got There , 2015, HotStorage.

[9]  Anastasia Ailamaki,et al.  Improving Flash Write Performance by Using Update Frequency , 2013, Proc. VLDB Endow..

[10]  Qi Zhang,et al.  Characterization of storage workload traces from production Windows Servers , 2008, 2008 IEEE International Symposium on Workload Characterization.

[11]  Bo Hong,et al.  File System Workload Analysis For Large Scientific Computing Applications , 2004, MSST.

[12]  Cheng Li,et al.  Assert(!Defined(Sequential I/O)) , 2014, HotStorage.

[13]  Alma Riska,et al.  Disk Drive Level Workload Characterization , 2006, USENIX Annual Technical Conference, General Track.

[14]  Jongmoo Choi,et al.  Improving SSD reliability with RAID via Elastic Striping and Anywhere Parity , 2013, 2013 43rd Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN).

[15]  Ajay Gulati,et al.  Storage Workload Characterization and Consolidation in Virtualized Environments , 2008 .

[16]  Antony I. T. Rowstron,et al.  Migrating server storage to SSDs: analysis of tradeoffs , 2009, EuroSys '09.

[17]  Jacob R. Lorch,et al.  A five-year study of file-system metadata , 2007, TOS.

[18]  Philippe Bonnet,et al.  uFLIP: Understanding Flash IO Patterns , 2009, CIDR.

[19]  Dutch T. Meyer,et al.  A study of practical deduplication , 2011, TOS.