Skip to content
November 16, 2012 / doganay


  1. No seperate SAN, disks are local and infiniband switches exist between exadata cells and disks.
  2. Using smart flash cache. (full rack->22 TB) Hot blocks held on these caches. For this reason x3 bandwidth and x20 read/write I/Os.
  3. Using smart scan. This means not all rows returned, only rows that match where predicate returns. So filtering is done not by instance, but by cell software.
  4. Using HCC (Hybrid Columnar Compression). This means data is compressed row based and column based. So, less I/O occurs.
  5. Using storage indexes: Min. and max. values of every column of every chunk (1 MB) is stored. So, where predicates processed faster.
  6. Using Flash Logging: Starts writing to redo flash memory and disk controller cache at the same time. Considers redo writing as finished when one of these finishes.
  7. Storage is capable to:
  • Join filtering (using bloom filter)
  • Incremental Backup Filtering
  • I/O Prioritization
  • Storage Indexing
  • Database Level Security
  • Offloaded Scans on Encrypted Data
  • Data Mining Model Scoring

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: