Data Loading Performance
Typical Issues:
- Loads are too slow to keep up with analysis needs
- Loading limited to nightly batch windows
Causes:
- Inability of loading to scale
- One node loading serially, instead of in parallel
- Co-dependent processing
- Queries and loading on same nodes
- Query performance degrades or causes downtime
- Your business cannot tolerate a query outage, delaying loading
- Destination bottlenecks
- One node receiving loads serially
- Data is queued up and bottlenecked by disk write speed of single node
- Network becomes a bottleneck
- Too much data being loaded through a congested interconnect
Remedies:
- Always-parallel performance and scalability
- On-demand incremental scaling of loading capacity
- Meet any loading service level agreement
- High-volume loading/exporting
- Loading and exporting scales linearly with additional nodes
- Loading/exports/backup is concurrent with query processing
- Massively-parallel, network-aware database architecture
- Leverage deep awareness of network topology
Top Picks
Whitepaper: New MapReduce Whitepaper |
Webcast: Bringing Big Data Analytics to the Enterprise - 11/12, with Merv Adrian |
Webinar: Service Oriented 'Analytics' - 11/19, with James Kobelius |
With Aster nCluster ... our data load time has decreased by over 95 percent, and our most important queries complete in seconds or less. ShareThis Tim Schigel, CTO |