The Vertica Analytic Database is a high-speed, relational SQL database management system (DBMS) purpose-built for analytics and business intelligence. The Vertica Database has a shared nothing MPP column-oriented architecture, and has been benchmarked by many customers as being on average 10x to 200x faster than other solutions. It also uses compression very aggressively, both of data on disk and data “in motion” during queries, which further enhances query speed while enabling cost-effective storage management. The Vertica Database runs on clusters of inexpensive, industry-standard Linux servers and requires virtually no setup and no manual tuning. Vertica offers three flexible deployment options: 1)install software on Linux servers of choice, 2)pre-configured and shipped on HP BladeSystem hardware and 3)licensed and used on an on-demand basis, hosted in the Amazon Elastic Compute Cloud (EC2).
VERTICA’S VALUE PROPOSITION
Reduce Costs: eliminate expensive hardware costs (runs on industry standard Linux hardware), significantly reduce data storage costs (up to 4 to 1 data compression capabilities) and on-going power consumption and cooling costs, radically improved price-performance - deploy large-scale query intensive databases for a fraction of the cost of other solutions.
Superior Database Solution: Magnitude faster performance (on average 10x to 200x faster) than other well known solutions. Painless scalability allows for scaling the size and usage of database without disruptive database platform changes, buying proprietary data warehouse appliance hardware and paying incremental DBMS software licensing fees for each CPU added to the system.
Increase Productivity: no setup and manual tuning – the Database Designer Utility recommends and applies the physical database design that provides the best data query/compression balance automatically. Faster performance and ease of use equates to better utilization of resource time and effort.
FINANCIAL SERVICES INDUSTRY EXAMPLE
Vertica saved the client over $1M and gave them better performance. This client had over 1.5 terabytes of stock market trade and quote history. The performance and cost wins in this application are typical for Vertica.
PROOF OF CONCEPT
Working in conjunction with HP, Vertica conducted a POC for Verizon Wireless. The goal was to measure several key performance indicators: flow detail record load times (insert rates), query response times, and compression ratios. Insertion, Query and Compression statistics from this POC illustrate the power of Vertica.
INSERTION RESULTS: Using 8 nodes to insert data (with 3 data streams per node), Vertica inserted 91,120 files (200k records / file) equating to 18,224,000,000 records inserted. The end result was an insertion time of 2 hours and 16 minutes equating to 2,249,876 records per second.
QUERY RESULTS: Testing 6 unique queries with 8 nodes per query with rows returned varying from 1 to 32 and total records in the database varying from 3,327,382,083 to 11,647,679,241, Vertica’s response time varied from .113 seconds to 3.2 seconds.
COMPRESSION RESULTS: Compression of 2.8 to 1 was achieved using a database with 3,327,382,083 records (only one day of records). The compression ratio is expect to be 3-4 to 1 as the database is scaled.
