Release Notes

Optimize an SC Series array for an Oracle data warehouse
20 Optimizing Dell EMC SC Series Storage for Oracle OLAP Processing | 2009-M-BP-O
For example, the following SQL statement creates a bigfile tablespace intended for storing the historical data.
The statement preallocates 1TB of space initially. When it is full, additional storage will be added in 1GB
amounts.
create bigfile tablespace HISTDATATS datafile '+DATA1DG' size 1024G autoextend on next 1024M maxsize
unlimited;
If a large amount of data will be loaded into the tablespace initially, preallocate enough storage to handle the
load and to improve the loading speed.
Choose an appropriate autoextend size to minimize the frequency of extending the data files and allocating
storage in the SC Series array during the regular data ingestion cycle. This provides the benefit of thin
provisioning and gradually allocating storage as needed without incurring excessive overhead.
3.4.6 Volume cache
Read and write cache generally improve performance for an all-HDD configuration. However, when SSDs are
used in the SC Series array, it is recommended to disable the write cache on volumes that use SSD storage
to improve performance. This means that any volume with a storage profile that writes to T1 storage such as
Recommended and High Priority should disable the write cache.
To enable or disable the read or write cache, in DSM > SC user preferences, enable the Allow Cache
Selection. See the Dell Storage Manager Administrator’s Guide for instructions.
3.5 Protecting Oracle with snapshots and a consistency group
Due to the large data volume in a data warehouses, backing up or restoring such an environment might take
days to complete, hence, affecting the organization’s ability to meet a specific recovery point objective (RPO)
and recovery time objective (RTO).
The SC Series storage snapshot feature is an ideal solution to address these challenges. Enabled through
the Snapshot Profile on each volume (see Figure 3), it protects an Oracle database by creating fast and
space-efficient snapshots of the storage volumes. A snapshot profile is a policy-based collection of rules
describing the type of schedule (once, daily, weekly, or monthly), a date/time and the interval to create the
snapshot, the volumes, an expiration time for the snapshot, and if the snapshot should be write-consistent
across all volumes in the schedule.
Guidelines for SC Series snapshots:
• Create a dedicated snapshot profile in DSM for each data warehouse/database.
• When a snapshot is taken, existing pages are frozen and no data is moved or copied. New data or
changed data are then written to new pages. Taking a complete snapshot of a data warehouse
usually takes seconds, regardless of the amount of data involved.
• Each complete set of snapshots represents a full point-in-time copy of the data warehouse.
• The snapshot size will depend on the amount of changes to the data since the last snapshot was
taken. Generally, it is significantly less than the original full size database.
• Thousands of snapshots are supported per volume.
• Data Progression moves frozen pages according to the selected storage profile to optimize storage.
See appendix A for storage profiles information.
• Snapshots offload the traditional backup processing, such as RMAN, from the primary server.
• They simplify the cloning of a full-size data warehouse for testing, support, or departmental use.