Gear6 Solution Brief: Electronic Design Automation

Introduction

Electronic design automation (EDA) applications create intensive data processing needs for the underlying storage infrastructure. In particular, many aspects of the EDA process such as regression testing and correction processing engines can lead to heavy peak loads. For designers intent on exploiting the full capabilities of these tools, optimizing performance and accelerating time to market, the entire system must be in balance.

The Challenge

While clustering processors has been a boost to performance at the CPU level, storage I/O operations have remained constrained by a reliance on mechanical disk, leaving the I/O infrastructure in need of attention. In particular, the storage bottleneck is exacerbated by the following factors:

Distributed processing

• Many EDA applications are distributed and run on dozens or even hundreds of clients simultaneously all accessing a single data set
• Concurrent access drives heavy workloads that can severely impact storage system performance due to the inherent latencies of mechanical disk

Workload scaling

• More complex designs drive higher application demands and can lead to I/O bottlenecks
• Common EDA scenarios that impact storage system performance include
• Many I/O requests for shared data leading to queuing
• High number of concurrent clients and transactions
• Support for millions of files
• Very large data sets

Service requirements

• EDA users strive for time to market and the ability to maintain consistent and predictable schedules
• Heavy peak loads can cause missed deadlines and project delays

Gear6 Solution

Gear6 addresses the need for predictable performance in EDA environments with centralized storage caching, and approach that enhances existing disk-based storage solutions with scalable, high-capacity caching appliances. This solution offers a number of benefits for EDA users:

Accelerate storage and applications

• Caching frequently accessed data in high speed memory delivers a huge boost in throughput and also access time to accelerate applications
• Reduced processing time speeds time to market

Reduce over provisioning:

• Centralized storage caching solutions delivers IOPS with high-speed memory instead of over-provisioned disk solutions
• Reducing excess capacity drives down power, space, and cooling costs, and simplifies infrastructure operation and maintenance

Improve Quality of Service

• Centralized storage caching solutions guarantee consistent and predictable response times for storage, minimizing peak load disruptions and ensuring timely product completion

Related Links:

• Centralized storage caching fixes the EDA compute bottleneck
• Shorten EDA Cycles With Storage Acceleration