Energy Exploration

The energy industry studies the earth with seismic data and models to accelerate resource discovery. Making models more accurate and faster pays off in reduced exploration costs and earlier time to market. The industry continues to develop new algorithms to increase accuracy, which depend upon ever-increasing CPU speed, expanding memory, and larger data volumes.
RNA MVX improves seismic models by speeding the movement of data and improving memory utilization. RNA MVX provides a Memory Cache capability to allow large data sets, such as travel time data, to reside in memory and be shared across the compute infrastructure. MVX also provides a Memory Motion capability that enables memory-intensive algorithms to address very large memory address space, backed up by an elastic and very fast swap space. The combination of these MVX functions can cut analysis and modeling times by 20X or more.
MVX for Seismic Modelling

MVX accelerated the customer's back testing by up to 20X with
no changes to the application
MVX provides a unique software only solution for accelerating large-scale analytic problems like seismic modeling. MVX leverages existing investments in network fabric and commodity servers, aggregating memory resources into a cloud that can be applied strategically to bottlenecks in the analysis process. MVX has several advantages:
- MVX builds on commodity hardware and standard operating systems.
- MVX does not require “fat nodes” with high-cost high-density memory
- MVX reaches beyond the compute cluster to marshal underused memory resources from around the data center
- MVX exploits high speed network interface card (NICs) and NIC features like remote direct memory access (RDMA) where it discovers them without requiring them on all nodes
- MVX provides an elastic memory cloud that can grow to be utilized anywhere in the data center, not just in the compute cluster
- MVX is up to 10X faster than SSD, and unlike SSD is a shared and re-deployable resource
MVX removes memory constraints to shorten turn around time on seismic processing jobs, while enabling more granular modeling with next generation algorithms. It does this without requiring costly server upgrades or additional layers of infrastructure. Energy businesses can extend their infrastructure life and add system scale at extremely low cost.



