Parsing a large computational process into smaller independent tasks that run in parallel to each other can help actuaries benefit from the time-saving efficiencies of cloud computing. Machine learning has parallel compute capabilities to assist with these tasks. In this article, Milliman’s Joe Long and Dan McCurley discuss how they were able to cut a three-month machine learning project down to four days using open source tools and the Microsoft Azure cloud.
Milliman today announced that it has been selected by Risk.net as the best tech vendor for actuarial modelling. For the second time in three years, Milliman is being recognized for its software solution Integrate™, which provides an innovative approach to support fully automated, end-to-end financial and risk modelling.
Integrate, which runs on the Microsoft Azure cloud, provides clients with best-in-class software for managing data input and complex modelling flows on the front-end, as well as robust business intelligence analytics and reporting capabilities of output with the recent addition of Microsoft Power BI Embedded. Additionally, Integrate provides a collaborative and governed process to support the model change process.
We are honored to have received this award as a recognition of Milliman’s innovative approach to client work, one that focuses on collaboration between experts in technology, actuarial, and risk solutions, and that drives value for not only the organization, but the policyholders, shareholders, and employees as well.
For more information on this year’s award, visit www.risk.net.
Integrate is a trademark of Milliman, Inc., registered in the U.S., U.K. and France. For more information about Integrate, click here.
Microsoft rolled out several Azure upgrades to improve its cloud platform’s performance. This TargetTech article quotes Paul Maher discussing how the upgrades can enhance Milliman’s Integrate® which runs on Azure. Here’s an excerpt:
Microsoft also released the L-series and H-series of VMs, which are tailored for high storage and CPU speed, respectively. These join other new instance types Microsoft has rolled out over the past year, including its GPU-based N-series and large instances for SAP HANA.
Milliman, an actuarial service provider based in Seattle, runs its platform-as-a-service offering on Azure and often runs different instance types to meet its clients varying needs.
“Matching the hardware and processing requirements at a price point and having a bastion of options is really important to us,” said Paul Maher, principal and CTO of life technology solutions at Milliman. “If you’re trying to run across a diverse client range, with varying price sensitivities, the last thing you want is a one-size-fits-all scenario.”
Another important upgrade for Milliman is the increased transparency from the preview release of Azure Monitor. The company wants to do more around diagnostics on the health of their workloads, and pass more feedback to customers about uptime, resiliency and performance analytics.
“For heavy users, which we are, it’s really important that we feel like we have granular control,” Maher said.
To learn more about Integrate, click here.
Milliman, Inc., the international firm of consultants and actuaries, today announced that MG-ALFA is capable of utilizing Windows Azure for on-demand processing. This represents a major step forward and a valuable feature as insurers look to expand computing capacity while controlling costs.
MG-ALFA Compute for Windows Azure, a proprietary computing component developed by Milliman in collaboration with Microsoft Corp., lets clients take advantage of a variable cost structure and gives them on-demand access to data centers with massive computing capacity. Running MG-ALFA on Windows Azure helps relieve insurers of the planning, management, and fixed costs associated with building and maintaining on-site computing resources. Infrastructure, support, maintenance, and monitoring of the grid are all provided by Milliman.
According to Van Beach, MG-ALFA product manager, “MG-ALFA Compute for Windows Azure was created from the ground up for high-volume actuarial computing. Our pilot companies are routinely running large-scale models using 10,000 cores provisioned in six Windows Azure data centers around the globe. This solution provides a scalable and highly available grid environment.”
Scott Guthrie, Microsoft corporate vice president, Windows Azure App Platform, says, “We are excited Milliman has chosen Windows Azure, just as it has been using Windows HPC Server since its inception. Moving MG-ALFA to the cloud is a great example of a scalable deployment on the leading edge of computing technology.”