AWS recently announced the second generation of Dense Compute Nodes - DC2 - for Amazon Redshift. (FlyData was the first data integration platform to report on the considerable performance boosts of the prior release in January 2014 and we’ll be excited to explore the new improvements ongoing.) As a primer, the type of node chosen on a Redshift cluster determines the CPU, RAM, storage capacity, and storage drive type for each node. Dense compute (DC) node types are compute-optimized whereas dense storage (DS) node types are storage-optimized. As such, the DC2 nodes are designed for DW workloads requiring low latency and high throughput. According to early reports from Amazon, customers can see higher query throughput and better concurrency with overall faster ETL jobs on the DC2 nodes. Amazon's statement that the new DC2 nodes will support I/O speeds of up to .60GB/s (compared to .20GB/s under the previous generation) is promising to improve customers' analytics performance. On a cost-basis alone, DC2 is a real boon, providing twice the memory per data slice and 30% better storage utilization due to layout optimization - all at the same cost of the prior generation's DC1 nodes. This translates into better ROI on the overall data analytics environment. If you’re continuing to use an RDBMS such as Amazon RDS, MySQL, PostgreSQL or Amazon Aurora, one of the long-term strategies will likely be to replicate traditional/relational analytics data stores into a cost-effective DW that scales at mass volumes. With the availability of DC2 nodes for Redshift, it's now even easier to realize the cost and performance enhancements by leveraging a data integration partner such as FlyData. By removing latency on back-end data processing with a tool such as FlyData for real-time database sync and replication, the new compute node improvements can further accelerate your BI time-to-value. FlyData Sync facilitates the BI goals of lean startups to large enterprises by replicating millions and billions of rows directly to Redshift in near-real time. We're excited to help our customers realize a wide variety of use cases where query speed is critical, such as in ad-delivery optimization and financial trading systems. As with the performance of Amazon’s new Dense Storage Compute nodes for Redshift, FlyData also continues to provide an evolving agile, intuitive experience to reduce unnecessary heavy-lifting and make your data fly! Stay tuned for more announcements on the way and be sure to find our team at AWS re:Invent 2017 for more details!
Snowflake vs Redshift: Which Cloud Data Warehouse is right for you? The Ultimate Guide to Redshift ETL: Best Practices, Advanced Tips, and Resources for Mastering Redshift ETL Learning about ETL - a founding engineer's personal account Redshift Unload: Amazon Redshift’s Unload Command The 5 V’s of Big Data