X

Everything you want and need to know about Oracle SPARC systems performance

Yahoo Cloud Serving Benchmark: SPARC S7-2 and Oracle NoSQL Advantage Over x86 E5-2699 v4 Server Per Core Under Load

Brian Whitney
Principal Software Engineer

Oracle's SPARC S7-2 server delivered 341 Kops/sec on 300 million records for the Yahoo Cloud Serving Benchmark (YCSB) running a 95% read, 5% update workload using Oracle NoSQL Database 4.0. NoSQL is important for Big Data Analysis and for Cloud Computing.

  • The SPARC S7-2 server was 1.9 times faster per core than a two-chip x86 E5-2699 v4 server running YCSB with a 95% read, 5% update workload.

Performance Landscape

The table below compares the SPARC S7-2 server and 2-chip x86 E5-2699 v4 server.  All of the following results were run as part of this benchmark effort.

 
YCSB Benchmark Performance
System Insert Mixed Load (95% Read, 5% Update)
Throughput
ops/sec
Ave Latency Throughput
ops/sec
Ave Latency Throughput
per core
Write
msec
Read
msec
Write
msec
SPARC S7-2
2 x SPARC S7 (2x 8core)
81,777 2.64 340,766 0.80 2.81 21,298
x86 E5 v4 server
2 x E5-2699 v4 (2x 22core)
155,498 1.38 502,273 0.46 1.16 11,415

Configuration Summary

SPARC System:

SPARC S7-2 server
2 x SPARC S7 processors
512 GB memory
2 x Oracle Flash Accelerator F320 PCIe card
1 x Built-in 10 GbE PCIe port

x86 System:

Oracle Server X6-2L server
2 x Intel Xeon Processor E5-2699 v4
512 GB memory
2 x Oracle Flash Accelerator F320 PCIe card
1 x Sun Dual Port 10 GbE PCIe 2.0 Low Profile Adapter

Software Configuration (for both systems):

Oracle Solaris 11.3 (11.3.8.7.0)
Oracle NoSQL Database, Enterprise Edition 12c R1.4.0.9
Java(TM) SE Runtime Environment (build 1.8.0_92-b14)

Benchmark Description

The Yahoo Cloud Serving Benchmark (YCSB) is a performance benchmark for cloud database and their systems. The benchmark documentation says:

With the many new serving databases available including Sherpa, BigTable, Azure and many more, it can be difficult to decide which system is right for your application, partially because the features differ between systems, and partially because there is not an easy way to compare the performance of one system versus another.  The goal of the Yahoo Cloud Serving Benchmark (YCSB) project is to develop a framework and common set of workloads for evaluating the performance of different "key-value" and "cloud" serving stores.

Key Points and Best Practices

  • The 300 million records were loaded into 4 Shards with the replication factor set to 3.

  • Four processor sets were created to host 4 Storage Nodes.  The default processor set was additionally used for OS and IO interrupts. The processor sets were used for isolation and to ensure a balanced load.

  • Fixed priority class was assigned to Oracle NoSQL Storage Node java processes.

  • The ZFS record size was set to 16K (default 128K) and this worked best for the 95% read, 5% update workload.

  • Sun Server X4-2L system was used as client for generating the workload.

  • The server and client system were connected through a 10 GbE network.

See Also

Disclosure Statement

Copyright 2016, Oracle and/or its affiliates. All rights reserved.  Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners. Results as of June 29, 2016.

Be the first to comment

Comments ( 0 )
Please enter your name.Please provide a valid email address.Please enter a comment.CAPTCHA challenge response provided was incorrect. Please try again.Captcha