Recent Posts by Brian Whitney

Using the Oracle Spatial and Graph option from the Oracle Database, one of Oracle's SPARC M8 processors achieved comparable performance to Oracle Server X6-2 system with two Intel Xeon Processor E5-2699 v4. The SPARC T8-2 server was configured with one SPARC M8 processor. The SPARC T8-2 server was configured using Oracle VM Server for SPARC to create a one-chip virtual machine (VM). The SPARC M8 processor showed a 1.8x advantage per core*seconds running sixty-four queries...

The STREAM benchmark measures delivered memory bandwidth on a variety of memory intensive tasks. Delivered memory bandwidth is key to a server delivering high performance on a wide variety of workloads. The STREAM benchmark is typically run where each chip in the system gets its memory requests satisfied from local memory. This report presents performance of Oracle's SPARC M8 processor based servers and compares their performance to x86 and IBM POWER8 servers. Bisection...

Oracle's SPARC M7 processor delivers best throughput per chip as measured by the newly-announced benchmark suite SPEC CPU2017. The SPARC M7 processor achieved world record per-chip scores: 114 SPECrate2017_int_base, 123 SPECrate2017_int_peak, 111 SPECrate2017_fp_base, 118 SPECrate2017_fp_peak. The SPARC M7 processor beat the Fujitsu PRIMERGY RX2560 M2 with Intel Xeon Processor E5-2699A v4 per chip by: 1.4 times on SPECrate2017_int_base 1.5 times on SPECrate2017_fp_base The SPARC...

Thanks to Karthik Ganesan for providing the majority of this content. The Java streams abstraction in JDK 8 provides an easy way to write and efficiently execute aggregate operations on large datasets. A key driver for this work is making parallelism more accessible to developers. In addition, Java Stream programs are often more concise, understandable and maintainable. Oracle's SPARC M7 and S7 processors can seamlessly make Java Streams programs 3x to 22x faster than x86. This...

Running the SPC-2 benchmark, the Oracle ZFS Storage ZS5-2 appliance using SSDs delivered SPC-2 Price-Performance of $8.89 and an overall score of 24,397.12 SPC-2 MBPS. The Oracle ZFS Storage ZS5-2 appliance delivered 1.5x improved performance and 1.4x improved price performance over the previous generation Oracle ZFS Storage ZS3-2 appliance as shown by the SPC-2 benchmark. Oracle holds four of the top ten price-performance results on the SPC-2 benchmark, more than any other...

Running the SPC-2 benchmark, the Oracle ZFS Storage ZS5-2 appliance with 12 trays of HDD delivered SPC-2 Price-Performance of $12.93 and an overall score of 19,610.36 SPC-2 MBPS. Oracle holds four of the top ten price-performance results on the SPC-2 benchmark, more than any other company. Oracle holds four of the top ten performance results on the SPC-2 benchmark, more than any other company. The Oracle ZFS Storage ZS5-2 appliance has nearly a 2.0x performance advantage over...

Running Yahoo Cloud Serving Benchmark (YCSB) against a 100 million record Cassandra database, Oracle's SPARC S7-2L server delivered 69,858 ops/sec running a 50% read, 50% write workload. Cassandra is used for Big Data Analysis and for Cloud Computing. The SPARC S7-2L server was 2.0 times faster per core than a two-chip x86 E5-2699 v3 server running YCSB with a 50% read, 50% update workload. Performance Landscape The table below compares the SPARC S7-2L server and 2-chip...

Processing lots of data with Java can require significant computing power. Oracle engineers have taken a two prong approach to improving Java performance. First, on general Java performance, Oracle's SPARC M7 and SPARC S7 processors provide up to a 1.5x performance per core advantage over x86 cores as shown by these benchmark results "SPECjbb2015: SPARC T7-1 World Record for 1 Chip Result" and "SPECjbb2015: SPARC S7-2 Multi-JVM and Distributed Results". This was achieved by...

Oracle's SPARC S7-2 server shows higher throughput performance, 1.6x per core, running a mixed transaction workload using Oracle Berkeley DB compared to  results on a two-chip, Intel Xeon Processor E5-2699 v3 server.  Each instance contains a sum of 50 million rows of customer, account and orders data.  The size of each instance is nearly 20 gigabytes in size. The SPARC S7-2 server delivered a rate of 365,195 transactions per second on the throughput test. On a per-core basis...