Real-Time Data Warehouse with MemSQL on Cisco UCS
We are at an inflection point where we are moving to a world where everything is connected to everything else. That’s what the Internet of Things (IoT) means. The IoT links objects to the internet creating the always-on economy by enabling data and insights never available before. By definition, the always-on economy requires real-time analytics. With analytics capabilities becoming the primary competitive differentiator, enterprises must now have the ability to analyze terabytes and petabytes of data, both in real-time as the data is being generated, and for business intelligence making use of past data. Increasingly, there is a need to join historical data with streaming data in real-time; something that has proven to be quite challenging as the size of the datasets have grown.
Analytics solutions have been a key part of our solution engineering portfolio from the very beginning when we introduced Cisco UCS back in 2009. This is true for both traditional relational-based systems, and emerging big data analytics (often described using the thee V’s: volume, variety and velocity; and sometimes with the three I’s as well: investments, innovations and improvizations).
Today I am happy to announce yet another addition to our growing portfolio of analytics solutions: Cisco UCS with MemSQL for real-time data warehousing applications.
MemSQL is a distributed, in-memory, relational database management system with full SQL compliance. It can ingest and transform millions of data events per day while simultaneously analyzing billions of rows of data using standard SQL as shown in the figure below.
Some of the key capabilities of MemSQL include:
- Fast data ingestion: Collect data using common message brokers such as Apache Kafka while maintaining durable, consistent delivery with exactly-once semantics
- Fast analytics: Query terabytes of data with advanced data compression using disk-optimized tables with high compression and vectorized queries for fast analytics
- Real-time analytics: Use memory-optimized tables to analyze real-time events
- Geospatial support: Store, query and index geographic data types, including polygons and points, to support area, distance and location analytics in real time
- JSON optimized: Store and query JSON data as a column type to efficiently store and analyze multi-attribute objects
- Fully distributed joins: Scale out fully distributed joins across any table and column for simple, efficient query access.
Cisco UCS Integrated Infrastructure for Big Data and Analytics is an ideal platform for MemSQL deployments – capable of processing high volumes of real-time or archived data, both structured and unstructured.
As shown in the figure below, our joint solution provides a scalable, real-time data warehouse platform for high-performance applications that require fast, accurate, secure and always available data, with linear scalability to millions of events per second while analyzing petabytes of data for insights.
The Cisco UCS Integrated Infrastructure for Big Data and Analytics with MemSQL provides a simplified, intelligent infrastructure and a real-time data warehouse with the scalability to meet growing business demands:
- Combines innovations from Cisco UCS such as programmable infrastructure with real-time analytics capabilities of MemSQL
- Designed and optimized for real-time analytics, internet of things, personalization and recommendations, risk management, monitoring and detection, and customer 360
- Pre-tested, pre-validated and documented by Cisco and MemSQL engineers to ensure dependable deployments that can scale from small to very large as workload demands
- Substantial business value for enterprise BDA deployments through scalability, performance, time to market, and cost effectiveness (Source: IDC).
The joint reference architecture is shown in the figure below. The architecture can scale as the workload demands, including expansion to thousands of servers through the use of Cisco Nexus 9000 Series switches.
For more information: