Oracle Training Oracle Support Development Oracle Apps

 
 Home
 E-mail Us
 Oracle Articles
New Oracle Articles


 Oracle Training
 Oracle Tips

 Oracle Forum
 Class Catalog


 Remote DBA
 Oracle Tuning
 Emergency 911
 RAC Support
 Apps Support
 Analysis
 Design
 Implementation
 Oracle Support


 SQL Tuning
 Security

 Oracle UNIX
 Oracle Linux
 Monitoring
 Remote s
upport
 Remote plans
 Remote
services
 Application Server

 Applications
 Oracle Forms
 Oracle Portal
 App Upgrades
 SQL Server
 Oracle Concepts
 Software Support

 Remote S
upport  
 Development  

 Implementation


 Consulting Staff
 Consulting Prices
 Help Wanted!

 


 Oracle Posters
 Oracle Books

 Oracle Scripts
 Ion
 Excel-DB  

Don Burleson Blog 


 

 

 


 

 

 

 

New Oracle TPC World Record

This new Oracle world record benchmark reached 110k queries per hour (QphH) (TPC-H uses large data warehouse queries) using a server configuration made up of a 64-Node HP ProLiant BL25p cluster, each with one Dual-Core AMD Opteron (model 280) processor and Red Hat Enterprise Linux 4, Oracle Database 10g with Oracle Real Application Clusters achieved a record-breaking performance of 110,576.5 QphH@3000GB with a price-performance ratio of $37.80/QphH@3000GB.

Here is the init.ora file that was used.  Vendors invest a fortune in bragging rights from benchmarks and you can learn quite a bit about how to tune your server kernel and Oracle parameters.  In this benchmark, note the large RAM regions (db_cache_size) and the setting of optimizer_index_cost_adj=1400, a value above the documented max of 1000, and likely implemented to enforce the large-table full-table scans required for TPC-H queries. 

--------- init_build.ora ------------
aq_tm_processes = 0
audit_trail = false
compatible = 10.2.0.1
control_files = (/home/oracle/dev/block/control1,/home/oracle/dev/block/control2)
db_block_checksum = false
db_block_size = 16384
db_cache_size = 2000m
db_file_multiblock_read_count = 48
db_files = 500
db_name = 10i
db_writer_processes = 4
dml_locks = 5000
global_names = false
instance_name = raca
log_buffer = 4194304
log_checkpoint_timeout = 1200
log_checkpoints_to_alert = true
max_dump_file_size = unlimited
nls_date_format = YYYY-MM-DD
open_cursors = 600
optimizer_mode = CHOOSE
optimizer_features_enable = 10.2.0.1.1
parallel_adaptive_multi_user = true
parallel_execution_message_size = 16384
parallel_max_servers = 100
parallel_min_servers = 48
pga_aggregate_target = 5500m
processes = 1000
recovery_parallelism = 8
replication_dependency_tracking = false
statistics_level = basic
undo_retention = 400000
optimizer_index_cost_adj = 1400
cpu_count = 2
shared_pool_size = 2000m

 

We also see the Linux Red Hat kernel parameters that were used in the benchmark.

 

--------- sysctl.conf ------------
# Kernel sysctl configuration file for Red Hat Linux
#
# For binary values, 0 is disabled, 1 is enabled. See sysctl(8) and
# sysctl.conf(5) for more details.
# Controls IP packet forwarding
net.ipv4.ip_forward = 0
# Controls source route verification
net.ipv4.conf.default.rp_filter = 1
# Do not accept source routing
net.ipv4.conf.default.accept_source_route = 0
# Controls the System Request debugging functionality of the kernel
kernel.sysrq = 0
# Controls whether core dumps will append the PID to the core filename.
# Useful for debugging multi-threaded applications.
kernel.core_uses_pid = 1
fs.file-max = 4194304
kernel.sem = 250 32000 100 128
#net.core.rmem_max = 524288
#net.core.rmem_default = 524288
#net.core.wmem_max = 524288
#net.core.wmem_default = 524288
#net.ipv4.ip_local_port_range = 32768 65000
#net.core.rmem_max = 2097152
#net.core.rmem_default = 2097152
#net.core.wmem_max = 2097152
#net.core.wmem_default = 2097152
#net.ipv4.ip_local_port_range = 32768 65000
#fs.aio-max-nr = 2097152
fs.aio-max-nr = 4194304
#kernel.shmmax = 4000000000
kernel.shmmax = 4563402752
#2.6 RKO
net.ipv4.ip_local_port_range = 32768 65000
net.core.rmem_default = 2097152
net.core.wmem_default = 2097152
net.core.rmem_max = 16777216
net.core.wmem_max = 16777216
net.ipv4.tcp_rmem = 4096 87380 16777216
net.ipv4.tcp_wmem = 4096 65536 16777216
net.ipv4.tcp_no_metrics_save = 1
net.ipv4.tcp_max_syn_backlog = 1536
net.core.netdev_max_backlog = 3000
net.ipv4.tcp_timestamps = 0
net.ipv4.tcp_sack =1
net.ipv4.tcp_window_scaling = 1

 



 

 
 
��  
 
 
Oracle Training at Sea
 
 
 
 
oracle dba poster
 

 
Follow us on Twitter 
 
Oracle performance tuning software 
 
Oracle Linux poster
 
 
 

 

Burleson is the American Team

Note: This Oracle documentation was created as a support and Oracle training reference for use by our DBA performance tuning consulting professionals.  Feel free to ask questions on our Oracle forum.

Verify experience! Anyone considering using the services of an Oracle support expert should independently investigate their credentials and experience, and not rely on advertisements and self-proclaimed expertise. All legitimate Oracle experts publish their Oracle qualifications.

Errata?  Oracle technology is changing and we strive to update our BC Oracle support information.  If you find an error or have a suggestion for improving our content, we would appreciate your feedback.  Just  e-mail:  

and include the URL for the page.


                    









Burleson Consulting

The Oracle of Database Support

Oracle Performance Tuning

Remote DBA Services


 

Copyright © 1996 -  2017

All rights reserved by Burleson

Oracle ® is the registered trademark of Oracle Corporation.

Remote Emergency Support provided by Conversational