Flash Memory technology, deployed as server-side PCIe or solid state disks (SSDs), is emerging as a critical tool for performance and efficiency in data centers of all scales. This presentation will discuss how the use of Flash impacts Cassandra deployments in terms of configuration, DRAM requirements and performance expectations. Ideas on leveraging C*'s cutting-edge data-center awareness to blend flash and disk storage nodes for cost and workload efficiency will also be shared. Flash media itself will be examined from a physical perspective to understand endurance issues. Data on write amplification under bulk-load and operational workload conditions will be presented to explain the impact to Flash of C*'s Log Structured Merge Tree architecture and the associated compactions. Finally, we will examine strategies to make Cassandra more Flash-aware using both conventional techniques as well as emerging Non-volatile memory (NVM) programming capabilities. Lessons learned from real-world customer deployments will be shared to complete this presentation.
3. #CASSANDRA13
KISSmetrics Customers Want
*Churn Prediction
*AB Tests
*Which Blog Posts and Ad Campaigns Attract High Value
Customers?
*User Conversion Funnel
*Revenue Prediction
*Customer Acquisition Costs
*Customer Lifetime Value
9. #CASSANDRA13
Where Secondary Indexes Break
Source: Place source content or footnote here. Delete if not needed.
*High Cardinality Data
*Only one index per query
*Indexes are distributed
*Only some datatypes; no counters
*Range queries are expensive
10. #CASSANDRA13
What Do I Want?
Source: Place source content or footnote here. Delete if not needed.
*Index high cardinality data; e.g. counters
*Complex queries, with multiple clauses
*Results in < 500ms for billions of rows
*Sub-field searching with regular expressions
*Range queries
35. #CASSANDRA13
Results So Far
*Results returned for an 8 clause query for 4 billion rows < 2
second
*Full regular expression support
*Full support for range queries
*Ability to index any numeric value, or value which can be
hashed.