Platform-as-a-Service for
Dynamic Parallel Processing

ThroughPuter News and Announcements

2019-June-11 ThroughPuter granted US patent 10,318,353 for concurrent program execution optimization:

ThroughPuter granted US patents 10,310,901 and 10,310,902 for dynamic resource management:

ThroughPuter launches alpha version of its demo-app -- for access, contact us via

2018-November-20 ThroughPuter granted US patent 10,133,600 for delegated scheduling and preemptive scheduling:

2018-November-20 ThroughPuter granted US patent 10,133,599 for application load and task type adaptive manycore processors and concurrent processing:

2018-August-28 ThroughPuter granted US patent 10,061,615 for application load adaptive concurrent computing technology:

2017-April-26 ThroughPuter granted US patent 9632833 "Scheduling application instances to processor cores over consecutive allocation periods based on application requirements":

2016-October-11 ThroughPuter granted US patent 9465667 "Application load adaptive multi-stage parallel data processing architecture ":

2016-September-20 ThroughPuter granted US patent 9448847 "Concurrent program execution optimization":

2016-August-23 ThroughPuter granted US patent 9424090 "Scheduling tasks to configurable processing cores based on task requirements and specification":

2016-July-26 ThroughPuter granted US patent 9400694 "Prioritizing instances of programs for execution based on input data availability":

2016-February-16 ThroughPuter granted US patent 9262204 "System and method for input data load adaptive parallel processing":

2015-October-6 ThroughPuter granted US patent 9152606 "Task switching and inter-task communications for multi-core processors":


ThroughPuter's presentation at 2015 HPC Advisory Council Conference:


ThroughPuter invited to present at the 2015 HPC Advisory Council Conference in Spain, on topic of Executing multiple dynamically parallelized programs on dynamically shared cloud processors:


ThroughPuter granted patents for manycore processors with hardware automated dynamic resource management functions, enabling multiple internally pipelined and/or parallelized applications to efficiently and securely share the parallel processing hardware:


ThroughPuter granted US patent 8935491 "Memory architecture for dynamically allocated manycore processor":


ThroughPuter granted US patent 8930958 "Efficient network and memory architecture for multi-core data processing":


ThroughPuter's presentation at FPGAworld 2014:


ThroughPuter invited to present at the FPGAworld conference (Stockholm, September 9):

(search for ThroughPuter)



ThroughPuter granted US patent 8793698 "Load balancer for parallel processors":


ThroughPuter granted US patent 8789065 for input data load adaptive parallel processing:


ThroughPuter granted US patent 8782665 for innovations enabling concurrent program execution optimization on dynamically shared multi-stage manycore processors:


ThroughPuter granted US patent 8769543 for application load adaptive time sharing of processor capacity:


ThroughPuter granted patents for application load and type adaptive manycore processor architecture:


ThroughPuter Announces Scalable Task-Centric Cloud Computing: Simpler System Implementation, Improved Application Performance:


ThroughPuter notes innovation milestones with issuance of US patent 8561078 "Task Switching and Inter-task Communications for Multi-core Processors" and US patent 8490111 "Efficient Network and Memory Architecture for Multi-core Data Processing System":


ThroughPuter wins patents for a cloud processor architecture:


Cloud Computing Journal article on open-source PaaS approach to parallel cloud application development and execution challenges:


ThroughPuter selected for high performance computing start-up showcase at SC12 (Provo UT, November 12, 10am at Adaptive Computing HQ):


GigaOM profile on ThroughPuter:

ThroughPuter's commentary - Regarding collaboration opportunities with other PaaS and IaaS providers: ThroughPuter's intent is to provide, either by itself or in partnership with other I/PaaS providers, a comprehensive parallel computing PaaS solution, to enable customers such as SaaS vendors to gain the performance and cost-efficiency benefits of dynamic parallel execution without the complexities of having to design and integrate their own parallel computing middleware and infrastructure. While such development front end to execution layer integrated parallel computing platform is what is needed to provide the necessary productivity for the application developers, for P/IaaS collaborators, ThroughPuter is prepared to offer just its dynamic parallel execution environment (based on the novel application load and type adaptive manycore processor architecture); the development environment is to be open-sourced and free for any party to host on any suitable infrastructure, allowing partners to customize and extend the PaaS for their distributions of the development environment, e.g. for particular application domains such as realtime big data analytics and business/operational intelligence, high frequency, algorithmic decision making, risk management, fraud detection/prevention, bioinformatics, engineering and scientific computing. Interested I/PaaS collaborators, please contact us at

View ThroughPuter's LaunchPad presentation:


ThroughPuter selected for GigaOM Structure 2012 LaunchPad (San Francisco, June 20):


ThroughPuter wins a patent for application program processing load adaptive processor cycle allocation:

ThroughPuter, Inc.

Website Builder