Design and Architecture of Dell Acceleration Appliances for Database (DAAD): A Practical Approach with High Availability Guaranteed

As IT organizations are pursuing database High Availability (HA) solutions to ensure and protest critical commercial data, the challenge is to leverage the three-fold key dimensions: cost, performance and availability. A successful solution needs to integrate Database Management System (DBMS) seamlessly with back-end storage and offer good performance and customer data protection. Through extensively investigating plenty of productions in the market, Oracle database is a leading solution for business critical application. As a cost effective solution, it could reduce business risk and achieve data availability, performance, and Return on Investment (ROI). On the other aspect, high-end storage solution, Storage Area Network (SAN) based on Fibre Channel protocol can be easy to deploy with Oracle database and match end-user's requirements. In this paper, we will present a design of high available database on a new reference architecture. The implementation of design is based on Oracle database with Dell Acceleration Appliances for Databases (DAAD), which introduces flash technology by Fusion-IO. To demonstrate practical performances, we built a proof-ofconcept platform and compared the platform with a traditional 96-SAS-drive platform. The results show that our approach can deliver more than 1-million random Input/output Operations per Second (IOPS) that is 27 times faster than the traditional platform and it can also achieve a 96% reduction in the latencies, thus showing the scalability of the approach with massive database nodes. Therefore, our approach provides not only an ultra-fast storage solution to boost database performances but it also offers a flexible and high available design to achieve zero-downtime database.

[1]  Hyojun Kim,et al.  BPLRU: A Buffer Management Scheme for Improving Random Writes in Flash Storage , 2008, FAST.

[2]  David A. Patterson,et al.  Latency Lags Bandwidth , 2005, ICCD.

[3]  Yuxiang Gao,et al.  Future Enterprise Computing Looking into 2020 , 2014, FCC.

[4]  Rina Panigrahy,et al.  Design Tradeoffs for SSD Performance , 2008, USENIX ATC.

[5]  Peng Zhang,et al.  Interlacing Bypass Rings to Torus Networks for More Efficient Networks , 2011, IEEE Transactions on Parallel and Distributed Systems.

[6]  Jae-Myung Kim,et al.  A case for flash memory ssd in enterprise database applications , 2008, SIGMOD Conference.

[7]  Alan Jay Smith,et al.  The performance impact of I/O optimizations and disk improvements , 2004, IBM J. Res. Dev..

[8]  Peng Zhang,et al.  Eigenanalysis-based task mapping on parallel computers with cellular networks , 2014, Math. Comput..

[9]  M. Malathi,et al.  Cloud computing concepts , 2011, 2011 3rd International Conference on Electronics Computer Technology.

[10]  Zaigham Mahmood,et al.  Cloud Computing: Concepts, Technology & Architecture , 2013 .

[11]  Sivan Toledo,et al.  Characterizing the Performance of Flash Memory Storage Devices and Its Impact on Algorithm Design , 2008, WEA.

[12]  Alan Jay Smith,et al.  Characteristics of I/O traffic in personal computer and server workloads , 2002, IBM Syst. J..

[13]  Kenneth Salem,et al.  Storage workload estimation for database management systems , 2007, SIGMOD '07.

[14]  Peng Zhang,et al.  Evaluation of Various Networks Configurated by Adding Bypass or Torus Links , 2015, IEEE Transactions on Parallel and Distributed Systems.

[15]  Mark Moshayedi,et al.  Enterprise SSDs , 2008, ACM Queue.

[16]  Günter Haring,et al.  Performance Evaluation: Origins and Directions , 2000, Lecture Notes in Computer Science.

[17]  Evan Marcus,et al.  Blueprints for high availability , 2000 .

[18]  Arif Merchant,et al.  A modular, analytical throughput model for modern disk arrays , 2001, MASCOTS 2001, Proceedings Ninth International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems.

[19]  Jim Gray,et al.  Flash Disk Opportunity for Server Applications , 2008, ACM Queue.

[20]  Sang-Won Lee,et al.  Design of flash-based DBMS: an in-page logging approach , 2007, SIGMOD '07.

[21]  Christian Engelmann,et al.  Symmetric Active/Active High Availability for High-Performance Computing System Services , 2006, J. Comput..

[22]  Sivan Toledo,et al.  Algorithms and data structures for flash memories , 2005, CSUR.

[23]  Louise E. Moser,et al.  Extended virtual synchrony , 1994, 14th International Conference on Distributed Computing Systems.

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

[25]  Nikolai Joukov,et al.  A nine year study of file system and storage benchmarking , 2008, TOS.

[26]  Christian Engelmann,et al.  Concepts for High Availability in Scientific High-End Computing , 2005 .

[27]  Xiaodong Zhang,et al.  Understanding intrinsic characteristics and system implications of flash memory based solid state drives , 2009, SIGMETRICS '09.

[28]  David A. Patterson,et al.  Computer Architecture: A Quantitative Approach , 1969 .