When building a system for real-time decision making, it is critical to optimize each element of the system architecture for balanced performance and long-term scalability. Today’s data velocity and variety can challenge conventional system technology, and unpredictable volume requirements add new complexities as customers consider how they will respond to tomorrow’s data generators. An imbalanced analytics system can result in bottlenecks that reduce the value of analytics tools and the enterprise data warehouses that power them. Most critically, a balanced infrastructure is key to ensuring the extraction of value to organizations at the time they need it—and within the budget they can afford. As such, it is critical for organizations to ensure that at either end of the ingest/egest process, they build scalable storage and storage processing technology that complements today’s high-speed data warehousing and data analytics technology.
DDN + Vertica: Real Life Results for Massively Scalable Big Data
Combining the industry’s leading big data analytics database with the leading data-storage solution for data-intensive computing environments, DDN and Vertica have delivered unmatched levels of data analytics capabilities. This joint solution combines the industry’s fastest analytics platform with the industry’s fastest storage engine. It can scale to ingest more than a terabyte of data per second from multiple, disparate data sources, including sensors, and it can process this data up to 60% faster than standard analytics tools and manage tens of trillions of rows of data with hundreds of billions of inserts per day.
The resulting solution can manage unprecedented amounts of data, scaling up to multiple petabytes, and process this data at unmatched speeds. More importantly, when you combine Vertica’s ability to lower disk space requirements with DDN’s ability to scale performance and capacity independently, you can deliver optimal performance with far greater efficiency.
This joint solution was recently validated by a mutual public sector client with extremely large data requirements. Read the joint solution brief to see how the user was able to achieve extremely fast time to actionable information from large datasets, compiled from hundreds or thousands of sensors, social media channels, and other data collection sources.
The Vertica Advantage
The Vertica Analytics Platform is the only database built from scratch to handle today’s heavy business intelligence workloads. In customer benchmarks, Vertica has been shown to manage petabytes of data and answers queries 50x to 1000x times faster than competing row-oriented databases and specialized analytics hardware.
Vertica’s SQL-based platform provides blazingly fast query performance for databases scaling from hundreds of gigabytes to multiple petabytes, running on industry-standard hardware, virtual machines, or a public cloud. What makes the Vertica Analytics Platform architecture so unique is that it combines the database industry’s most significant recent innovations into a single database for the first time.
- Columnar database architecture
- Distributed, grid computing, and shared-nothing architecture
- Aggressive data compression
- Automatic physical database design
- Automatic “log-less” recovery by query
- Scale-out MPP architecture on industry-standard servers, virtual machines, or on the cloud
- Hybrid transaction architecture that supports concurrent, parallelized querying and loading of data
- Multiple physical sort orders and joins of related datasets (“projections”)
- Native high availability without hardware redundancy
- Universal connectivity via JDBC/ODBC and SQL to BI applications, ETL, and querying/ reporting tools
The DDN Advantage
The DDN Storage Fusion Architecture® (SFA) combines unprecedented IOPS and bandwidth performance with highly efficient capacity management to maximize application performance and minimize overall total cost of storage system ownership for data-intensive environments. The scalability and performance of DDN storage systems, including the fastest storage systems in the world with aggregate performance beyond one terabyte per second, helps organizations maximize the value of information and accelerates analysis and critical decision-making.
The DDN solution supports storage arrays, file systems and object storage appliances for the Cloud to the world’s most data-intensive environments. These scalable, highly efficient storage solutions enable our customers to accelerate time to results, scale simply as data sets continue to grow and gain competitive advantage through resolving performance and capacity scaling challenges. By optimizing each element of the I/O environment for performance, capacity, and data center efficiency, DDN solutions deliver the highest levels of performance and fastest time to results.
- 800% faster than competitive solutions – single systems provide up to 40GB/s, and reaching 1TB/s can be achieved in just 25 systems
- Enormous internal bandwidth – the DDN architecture protects performance during drive rebuilds, while competing systems lose as much as 40%
- In-Storage Processing® – enables latency-sensitive applications to live right inside the storage appliance
- ReACT® intelligent cache management – optimizes and writes in real-time: sequential data goes directly to disk media while small, random IO utilizes extremely fast cache
- Read IO quality of service – read IO doesn’t suffer due to a single, unresponsive disk
- DirectProtect real-time error detection and correction – increases data resiliency and reliability with little performance impact
- Storage Fusion Fabric® unparalleled back-end SSD support – fully utilizes SSDs for unprecedented levels of sustained random IOPS
- Up to 1TB/s ingest speed
- Industry-leading solution scalability
- Unmatched query speed
- System flexibility and ease of management