Predict and Optimize Cluster Performance
Optimize and plan your big data cluster performance and network behavior with Intel® CoFluent™ technology for big data.
Intel® CoFluent™ technology for big data is a planning and optimization solution that predicts cluster performance and network behavior for big data challenges. Intel® CoFluent™ technology for big data helps address common big data cluster design challenges: predicting system scalability, sizing the system, determining maximum hardware utilization, minimizing costs, and predicting system performance. This technology allows you to optimize and plan according to your business needs, and provides a solution to help you minimize IT spending on big data clusters.
Highly configurable and with the ability to simulate clusters in software environments, this technology eliminates the need to set up cluster applications just to identify resource requirements: Determine optimal cluster size; identify optimal storage, I/O, and compute resources; scale up strategy; scale out strategy.
Plan and optimize your big data cluster for faster discovery and quicker resolution of performance bottlenecks: Avoid system over-provisioning; guarantee optimal software settings and performance; maximize ROI.
Dramatically reduce the costs and time it takes to switch from a PoC stage to a scalable production deployment: Predict hardware utilization and software states; predict job execution times against hardware and software changes; accurately simulate hardware, software, and workload variations.
Learn how to meet real-world challenges using Intel® CoFluent™ technology for big data.
Challenge: Create a simulation model for a video streaming system that optimizes the usability and stability of a network during deployment. This includes estimating the network, disk, and node requirements to support 1,000 concurrent users, and identifying the average throughput rate.
Solution: Based on a baseline system simulation model, use an Intel® CoFluent™ technology for big data simulation model to determine the node count requirements; identify possible network bottleneck issues and possible solutions; and determine optimal system disk options.
Challenge: Determine the optimal hardware and software configurations for typical camera-based use scenarios.
Solution: Use an Intel® CoFluent™ technology for big data simulation model to estimate the number of nodes required to support 1,000 cameras based upon insert intensity, query intensity, and a balanced scenario of intensities. The simulation model also estimates the ratio of HD cameras to server with Intel® Xeon® processors required for the video analytics system.
Challenge: Estimate the configuration requirements to complete a sort for a 60 GB file within 200 seconds.
Solution: Use Intel® CoFluent™ technology for big data to build a simulation model that can determine the lowest cost configuration with a minimum number of hardware/software components to meet the 200-second execution time.
Challenge: Optimize offline banking, including storage, compressing, partitioning, and caching processes.
Solution: Use Intel® CoFluent™ technology for big data simulation model of reporting and deep analytic query processes. This includes making use of latency, throughput, hardware utilization, software status, job- and application-specific data, and custom metrics.
Success Brief: Siemens develops telecom designs using Intel CoFluent™ studio,...Preview | DownloadView (685.5 KB)
Success Brief: Sensata Technologies creates digital imaging performance models,...Preview | DownloadView (632 KB)
White Paper: Shares lessons learned for evaluating real-time processing...Preview | DownloadView (776 KB)
See how Intel® CoFluent™ Technology for Big Data and Intel® Rack Scale Design...Preview | DownloadView (1.1 MB)
Get support for Intel® CoFluent™ studio and Intel® CoFluent™ reader products. Access your self-service portal or submit new cases to customer service.
Submit an inquiry to get information on Intel® CoFluent™ technology and the university program, or to request documentation and downloads.