Unleash the full potential of your IT infrastructure
Origami provides a cost-effective framework which is easy to use and maintain, robust, and highly flexible – ensuring quicker delivery of the right answers, and a significant improvement on HPC ROI. Origami was developed to utilise parallel calculation to deliver benefits across any HPC grid environment.
A cost-effective framework alternative: robust & highly flexible
Xifintiq have partnered with NAG – the market leading numerical library and tools provider to investment banking – to launch Origami. NAG's experience in rigorous testing and correctness combined with its respected position in the quant finance community deliver an unrivalled level of service to Origami users. This service includes tailored implementation and installation ensuring users gain maximum benefit from the framework, user training, plus first-line technical support.
- Origami puts the power of on-premise, cloud and hybrid infrastructure resources at your fingertips
- Origami removes constraints that are inherent in other grid solutions and empowers you to deliver faster solutions
- Origami delivers optimised workload scheduling using Directed Acyclic Graphs (DAGs), and performance scales whatever the size of the grid
- Origami offers game-changing robustness and flexibility
- Origami reduces cost by enabling simultaneous easy selection of heterogenous hardware (CPU/GPU)
- Origami lets you identify and correct compute bottlenecks
- Origami provides the ability to predict the required execution time and cost for tasks performed multiple times
Empower faster solutions
- Origami has fewer constraints than other grid solutions which empowers you to deliver faster solutions
Deliver competitive advantage
- Origami reduces task distribution time thus allowing you, and your business, more time to focus on other critical business activities
- Origami lets you identify and correct compute bottlenecks in a quick and easy way – more accurate results
Origami is fully maintained and supported by world-class industry experts – future proofing your investment
Origami provides a combination of tools that allows the user to quickly set up their own grid, create tasks and start distributing them across your HPC environment. Users can either define their requirements as a series of individual tasks or as linked interdependent tasks.
A Directed Acyclic Graph (DAG) is a series of dependent tasks where the dependencies may be non-trivial and there are no circular dependencies. Origami offers a DAG Manager that takes a DAG definition and optimises the distribution of the tasks across the Origami grid, providing both timing and cost metrics alongside the results. The DAG Manager comes with the ability to restart a DAG job, picking up where it previously left off.
For data distribution there is the Hierarchical Group of Vectors (HGV) data format designed to enable programmers to easily and quickly build the data structures they need. HGVs can store labelled scalar, vector and tabular data of different types along with their relationships in a natural, spreadsheet-like fashion. For the web the data can be streamed using compression, to save space and time, and encryption to protect your data in a web friendly way. Also provided is an 'App’ that allows you to visualize and modify your HGV files.
Origami in the cloud is designed to grow with you. You can control how and when tasks are scheduled out of your internal environment into your hosted cloud service, be that Microsoft Azure, Amazon AWS or Google Cloud. Once your cloud service provider is chosen, configuration is simple. You switch from in-house architecture to cloud by changing your connection string. Origami will estimate the execution time and costs prior to starting any execution and provide full actual metrics and costs upon completion.
For DevOps testing and bespoke HPC work the user can create individual tasks that can be simply sent to the grid.
Origami can be installed on networked machines to offer an in-house grid, or used as an execution platform with cloud providers. Currently Microsoft Azure, Amazon AWS and Google Cloud, and with other providers in the future. Once set up, switching from in-house to Cloud is as simple as changing a connection string. Thanks to its integrated runtime metrics Origami provides the ability to predict the required execution time and cost for tasks performed multiple times prior to execution.
Origami can quickly be installed on your local network of machines. Origami maintains one or more queues upon which client applications can place requests. Individual nodes (machines on the grid) can accept tasks from one or more of these queues depending on their specialist hardware (e.g. GPU), compute hardware (multiple cores) or availability (for example dedicated server versus development workstation). The Origami Grid Controller allows monitoring and control of applications running on the grid. As your needs grow, the grid can easily expand to include cloud services. All of this can be accomplished without the need for specialist HPC support personnel.
Supported platforms are Windows (7+), Linux and MAC with programming environments .NET and C++.