Appistry's Ayrris Computing Environment

To realize the promise of individualized, personalized, next-generation medicine, research hospitals and medical systems need to think beyond one-off black boxes, pay-as-you-go virtual infrastructures, or unwieldy systems like Hadoop.

Appistry's Ayrris™ is a computing environment that enables the easy assembly of big-data solutions in a fraction of the time and cost typically associated with big-data projects. Leveraging the experience acquired over a decade of implementing mission-critical, data-intensive analytics applications in industries like finance, logistics, and defense, Appistry built Ayrris to

• Be easy to set up, run, and maintain. Extending infrastructure is as easy as plugging in a server—and Ayrris can run on any hardware.
• Accommodate any analytics, as is—critical for major research hospitals and medical systems that have invested heavily in developing analytics pipelines, recipes, or workflows for bioinformatics, genetic analysis, and data integration and manipulation.
• Provide life cycle support for big, unstructured data, from collecting it off sequencers or from other medical systems and diagnostics through processing to transferring it into applications and systems for interpretation and storage.
• Offer a variety of deployment options. Access Ayrris on your infrastructure or ours, get it on a dedicated appliance, or burst into an Ayrris cloud to gain added capacity or throughput when you need it most.

What is Ayrris?

Ayrris™ is a big-data computing environment designed for organizations that want to capitalize on the results of analytics rather than spending their time, money, and resources on the processes needed to run analytics at scale. Sitting between systems generating large volumes of unstructured data (such as sequencers, medical diagnostics, or EMRs) and applications for interpreting and storing data, Ayrris weaves together compute power with an organization's preferred analytics to transform data into actionable information.

Ayrris comprises the following components:

• Fabric computing technology that orchestrates any collection of commodity machines into a single fabric of fully autonomous, cooperative workers that has no single point of failure and provides management simplicity—add or subtract workers or update configurations without taking systems offline or disrupting analytics.
• Computational Storage™, which distributes both computational processes and data across all nodes in a fabric to minimize data movement and maximize data center capability and throughput.
• An analytics pipeline framework that can accommodate any type of data abstraction. Ayrris includes over 60 open-source and commercial NGS analysis tools as well as general analytics algorithms such as MapReduce, sparse-matrix multiplications, scatter/gather, and Monte Carlo analysis. With Ayrris, organizations can create and run powerful analytics pipelines without complex programming.
• Pipeview, an interface to the analytics pipeline framework that is used to assemble and run pipelines and monitor their execution.

Benefits of Ayrris

Hospitals and medical research institutions don't have the time or money to engage in full-press IT projects. They need a big-data computing environment that can assemble the underlying compute and analytics to run their unique science reliably, easily, and cost-effectively. Ayrris™ delivers the following benefits to organizations seeking to transform big data from a management headache to an actionable asset:

Maximize infrastructure investments		Unprecedented core efficiency
Ayrris requires no specialized hardware to run, so you can install it on any machine—from a commodity server to a supercomputer. And whatever your infrastructure choice, Ayrris will turn it into an optimally efficient fabric that is easy to maintain.		Computational Storage™ enables every node in an Ayrris fabric to serve as both a storage device and compute engine, minimizing data movement to reduce I/O wait times and maximize resource allocation.
Maximized throughput		Rapid pipeline and application development
An Ayrris fabric allocates resources as needed for each step in a job rather than reserving resources based on a maximum amount of compute needed for any given step, enabling it to accomplish more work than traditionally architected systems.		Pipelines are easy to construct and run with Ayrris because there's no need to rewrite algorithms to fit a specific data abstraction or inflexible API. And because Ayrris focuses on inputs, outputs, and executables, it streamlines application development to weeks rather than months or years.

Why Ayrris for Big Data?

The end-products of big-data management can drive business decisions and set a company apart. But producing these end-products is costly and time-consuming for research hospitals, medical systems, and diagnostic labs whose core competency is not (or shouldn't be) informatics.

Download a white paper to learn about the four capabilities in a big-data environment that help organizations bring their preferred analytics into product extensibly and cost effectively.

• Adaptable architectures capable of meeting ever-increasing compute and storage demands
• Ways to minimize data movement and maximize data center capability and analytics throughput
• A flexible analytics framework that can handle any type of data abstraction so that organization can advance their workflows, recipes, or pipelines unchanged
• Reliable, cost-effective deployment options that keep data secure and accessible to all and only those who need it.

You'll read how Appistry utilizes Ayrris™ to deliver these capabilities and helps organizations solve the big-data problems impeding next-generation medicine—faster and more cost effectively than other technologies.

 

Download the white paper now!