Database compression isn’t new, but
Storage Optimization automates key
tasks to make it easier to use. For
example, it builds the compression
dictionary automatically, instead of
requiring the DBA to manually specify
the patterns to be compressed.
Fred TX
10000 500
Plano
20000 Plano 500 24385 John TX
Fred
10000 (01)
(02)
… (02) 20000 (01) John
Informix also provides guidelines
ee sidebar, “Resources”) for when
optimization would be least useful,
such as for small tables that fit into
memory without compression and for temporary tables and
system catalogues. But even in those cases, the CPU cycles
needed to compress the data and the table scans required to re-
pack and shrink data have little impact on overall performance.
c
c
c
(s
Figure 1: Saving repeating patterns as shorter character strings allows Storage Optimization to
achieve very high compression ratios.
While compression and decompression do require CPU
ycles, “We’re finding the price you pay in CPU overhead is
outweighed by the benefit you get from having fewer pages
and fewer I/Os,” says IDS Chief Architect Kevin Brown. “If
before, your database couldn’t fit in memory, and now it can,
you’re going to see a really big performance boost.”
Feature V11.50 includes the SQL Warehouse (SQW) Client
with Design Studio for data modeling, schema design, data
transformation design, and data flow design; SQW Warehouse
Server, with an administration console to schedule and manage data flows; and SQL Warehouse runtime to perform data
transformation within the IDS data server.
The Informix Warehouse infrastructure also supports
ntegrated external tooling for BI analytics from IBM
Cognos, for managing data growth with IBM Optim, and
for data transformation and cleansing with IBM InfoSphere
In cases where compression might compromise appli-ation performance, administrators can specify which fragments to compress. “If a table has 10 fragments, for example,
you might want to leave the most recently created fragments
uncompressed for performance purposes, and compress
the older fragments, since they’re less likely to be accessed,”
says Brown. A recompress option allows administrators to
rebuild the compression dictionary to ensure optimal compression of the tables, which is useful if the data within them
has changed significantly over time.
d
p
i
DataStage and InfoSphere QualityStage.
With the new IDS data warehousing tools, DBAs can
erform business analysis on their operational data stores
to ensure access to the most current data. They can also
create separate data warehouses within IDS to prevent analytic queries from slowing production applications or run
a “shared disk” configuration in which separate servers
access the same data for OLTP and business analysis.
DBAs can use any or all of the Storage Optimization
apabilities while the database is running, thus preventing interruption to either transactions or queries. They can
also manage Storage Optimization from within the same
IDS OpenAdmin Tool (OAT) they use to manage the size of
volumes and to monitor performance. Graphical tools show
administrators or business users how much space they
can save by running any of the functions within Storage
Optimization. Storage Optimization can also be controlled
through a command-line interface, which is useful for the
many ISVs that bundle IDS within their own applications.
The bottom line: businesses that have been using IDS for
ata warehousing and BI now have better, more targeted tools
to achieve their goals. “We give organizations the choice so they
can make the best use of their data and infrastructure based on
their technical and business requirements,” says Brown.
RESOURCES
… 24385
Dept 500
Plano, TX
24385
… …
(01)
(02)
Bob Scheier is a veteran IT writer based in Swampscott,
Massachusetts.
The benefits of the bundle
Storage Optimization is available in a new data warehouse
platform package called IBM Informix Warehouse, which
includes IDS, its new Storage Optimization Feature, and the
IBM Informix Warehouse Feature V11.50. Informix Warehouse
Data warehousing with Informix: ibm.com/informix/warehouse
IDS Storage Optimization: http://download.boulder.ibm.com/
ibmdl/pub/software/data/sw-library/informix/whitepapers/
compression.pdf
IBM Informix Warehouse: ibm.com/common/ssi/rep_ca/2/897/
ENUS209-142/ENUS209-142.PDF
IBM Informix Warehouse Feature: ibm.com/common/ssi/
rep_ca/1/897/ENUS209-041/ENUS209-041.PDF