Innovating with EDB Postgres Plus Cloud Database

June 04, 2015

Postgres Plus Cloud Database (PPCD) customers have more flexibility in how their applications will interact with their database than other rigid, black-box database services.

Postgres Plus Cloud Database (PPCD) deploys Postgres databases in private instances that customers can configure and customize in any way they want. And many of our customers do just that. They add plugins to the database, make use of compute-heavy stored procedures, and even place other applications into the database instances to achieve higher performance at a cost savings – because they don’t need a second, separate instance to run the application.

That flexibility allows our customers to build innovative operational architectures for their application and database. And now, PPCD supports a broad range of new generation Amazon instance types that provide even better performance for customer systems.

The newest release of EDB Postgres Plus Cloud Database [PPCD] offers support for all new-generation classes of virtual machine instance types. Amazon has made enormous strides in not only lowering the price of their instances, but also in creating classes of instances that excel with specific and sometimes non-traditional database workloads. You can read about them on Amazon, but I’ll summarize them here so you can select the best instance type that will optimize your innovative Postgres workload.

For no frills, less intensive database workloads, T2 and M3 instance types offer good all-around performance. You’ll find these offered by other databases on AWS.

But if your application needs, for example, more CPU performance or lots of in-memory processing, you have other choices that will provide better performance with PPCD. Read on for details.

Analytical Application Options

If you have lots of stored procedures, or have chosen to run analytics, scientific or other high-performance applications in your private PPCD instances, C3 and C4 instance types will give you significantly greater performance. If you are using PostGIS for route calculation, nearest neighbor, large radius distance joins, or other computational tasks, you should definitely consider C3 or C4 compute optimized instances.

Memory Optimized Options

R3 memory optimized instance types are perfect for high-performance databases like Postgres. If your database fits in shared_buffers, then your reads will probably be very fast. Additional RAM available to the database will be used for sorts, hashing, and materialization (work_mem) or vacuuming and index builds (maintenance_work_mem) – all of which will benefit from R3 SSD-backed instances. Moreover, R3 instances feature the lowest AWS price point for GB of RAM.

Storage Optimization Options

Storage optimized I2 instances deliver fast SSD-backed storage with very good random I/O performance. Another kind of storage optimization is provided by D2 dense-storage instances. Users who wish to place high-performance or parallel processing applications on the same instance as the Postgres database will find that I2 instances offer high IOPS at a low cost and D2 instances offer the lowest price per disk throughput processing on Amazon EC2.

Parallel Processing Options

Finally, although G2 GPU instances are typically associated with graphics processing, if you require high parallel processing capability, you'll benefit from using the G2 GPU instances. They provide access to NVIDIA GPUs with up to 1,536 CUDA cores. You can use GPU instances to accelerate many scientific, engineering, and rendering applications by leveraging the Compute Unified Device Architecture (CUDA) or OpenCL parallel computing frameworks.

EDB Postgres Plus Cloud Database Offers the Widest Range of Choice on Amazon

Whether your PPCD private database instances need better computing performance, higher performing on-instance disks, faster memory, or accelerators for parallel processing, Postgres Plus Cloud Database supports the Amazon instance type you need.

Share this