Gpu digital lab investors presentation

39
GPUDigitalLab Project Manager Gubanov Oleg Igorevich

Transcript of Gpu digital lab investors presentation

PowerPoint

GPUDigitalLabProject ManagerGubanov Oleg Igorevich

1

GPUDigitalLabAim of The Project:

To provide access to parallel computations for scientists and lab workers at a reasonable cost.

, . , . , 32

2

GPUDigitalLabSolution We, the members of the Axioma Software team, would like to purpose a cluster solution for parallel computations on the GPU. This product will consist of a GPU oriented server that will contain NVIDIA Tesla Graphics Processor at its core. The software would be built upon Microsoft DirectCompute engine. It will be built as a set of client applications that use the power of the GPU core for the computtations. Each application would be oriented to either a problem or a set of problems in modern science and computer graphics.User starts by logging into the server and download the relevant client application. After that the user fills in an input form and sends the data to the server through a secured channel. This architercture allows users to use the power of modern gpu despite the fact they have relatively cheap hardware., . , . , 32

3

This project consists of a gpu processing core engine that has a set of connected client applications working in allocated domainsThis project has a scalable architecture that makes it easy to install new products.The aim of the project is to provide the scientific community with a powerful computational platform at a reasonable price.The website of the project includes a dedicated control panel for each user where he can see the current account balance as well as the list of the latest operations.

Project OverviewGPUDigitalLab

Software Architecture

, . , . , 323D Graphics Core EngineDirectCompute Core EngineVideo Rendering EngineDirect2D Graphics EngineCore EngineFluid Mechanics Rendering EngineData Visualization engineFPS Scene Rendering EngineRender Farm Engine3-rd Person Simulations EngineMathematical Modelling Engine

GPUDigitalLab

Software ConceptAt the core of the system there is module that can execute compute shader programs and analyze resultsThere are 3 types of data that we frequently need for our purposesStructured Buffers(used to store numerical data)Shader Resources(used to store texture dataUnordered Access View(used to send the collected data to the computational pipelineCompute Shader(a module that collects the data stored in buffers and performs computations based upon a certain algorithm

Program StartupOn startup the program open a login dialog

Login

Password

Program RuntimeAfter Logging in the system creates a user session and sets it a unique id. Using the locking mechanism of compute shaders we create a set of writable buffers, shader resources and UAVs(unordered access views).The system loops through the .config file and creates the execution domains for every core module.

Program RuntimeIn order to run client applications within our core we need the following objects for each applicationApplication Manager(responsible for launching and shutting down apps).Application Instance(responsible for controlling the app execution thread. It must collect the data produced by the apps).Event Processor(responsible for handling the messages produced by the client apps and processing possible errors)

Program Startup

Direct3D Initialization

DirectCompute execution process

Application Domain hasAn initialized 3D Rendering LoopAn initialized DirectCompute processing loopA set of buffers for data storageA set of shader resources for texturingA set of compute shader instancesAn allocated DirectCompute manager class for operations such as data creationAn allocated Data archiving module for compressing and decompressing data.

Application Domain ManagerCreates and destroys DomainsCollects the data from event processorsKeeps the diary of the operations.Controls the threads that are used by the domain

Application Domain InstanceHolds the objects that are necessary for computationsHas a collection of program objects such as buffers, resources and views.Provides a mechanism to edit the data stored in buffers.Provides a secure access to the data for client apps

Application Domain InstanceAn allocated memory pool for application executionContains a set of predefined objects, buffers and resources.Allows to transfer data securely between different processes.Allows to load program utilities into its threads and control the operation

User Session ControllerProvides the user with a secure access to system resourcesCreates a session with a unique session id and stored its in a data archiveStarts a thread that processes the actions of the user and sends the results to the system modules

Application ManagerHas an id of a running software processControls the data that is produces by the processResponsible for starting and terminating systemic widgetsResponsible for transferring the data between widget.

Application Event ProcessorControls the event produced by the application through a named pipe and an allocated reading threadUsed the received data to determine the state of the executed applications.Sends the received info about an application to application state manager

Application State MangerResponsible for collecting the data from application processors about the state of a moduleResponsible for informing the other participating modules about a state change for a given module.Responsible for sending the data about the application errors to the main processing loop.

Program typical execution thread

Cluster products of GPUDigitalLab, . , . , 32

GPUDigitalLab

7 Steps to Use GPUDigitalLab, . , . , 32GPUDigitalLab

The examples of GPUDigitalLab ProjectsFluid Mechanics, . , . , 32GPUDigitalLab

Chemical Reactions Simulations

, . , . , 32GPUDigitalLab

Blood Circulation Simulator, . , . , 32GPUDigitalLab

Crowd Rendering Simulator, . , . , 32GPUDigitalLab

Ray-Tracing Rendering System

, . , . , 32GPUDigitalLab

Computational FLUID MECHANICS

More fluid mechanics

FRACTALS ENGINE

Upcoming ProductsGPUSmartCrowdEngine software to visualize and classify crowds of people for statistical analysisGPUProcessAccelerator system utility that allows to transfer processing threads of data from cpu to GPUGPUVideoInspector software to seek relevant text and numerical information inside a video fileGPUDMOLSimulationEngine software products for molecular configurations computation and dispertion of the elextron density.GPUSkinInfectionDetector software product that uses image analysis for detecting skin diseasesGPUConvectionVisualizer software to visualize air streams within an apartment buildingGPUFireExtinguishingPlanner training tool for a fire brigade or the workers of a factory where you can configure the interior of the apartment, set random fire sources and create a training scenario. A group of students should eliminate the fire during a limited amount of time.GPUConstructionDemolitionEngine building destruction simulation engine.

, . , . , 32

Upcoming ProductsGPUChemicalReactionsSimulator a learning game where students have to construct a chemical reaction equation using an interactive periodic table.GPUBloodSimulationEngine blood circulation engine.GPUCavitiesSimulationEngine dental diseases simulation engine.GPUFlueAndColdSimulationEngine cold and flue dispersion simulator.GPUCrudeOilFlowSimulationEngine oil pipe traffic simulation engine

, . , . , 32

Essential HardwareServerModel: GPX XT10-2260-6GPUCPU: 2 x Six-Core Intel Xeon Processor E5-2630 v2 2.60GHz 15MB Cache (80W)RAM: 8 x 4GB PC3-14900 1866MHz DDR3 ECC Registered DIMMHDD: 250GB SATA 6.0Gb/s 7200RPM - 2.5" - Seagate Constellation.24 x 800GB Micron M500DC 2.5" SATA 6.0Gb/s Solid State Drive2 x 1.6TB Intel DC S3500 Series 2.5" SATA 6.0Gb/s Solid State Drive2 x 800GB Intel DC S3700 Series 2.5" SATA 6.0Gb/s Solid State DriveGPU: NVIDIA Tesla K40M GPU Computing Accelerator - 12GB GDDR5 - 2880 CUDA CoresNetwork Card: Intel 10-Gigabit Ethernet Converged Network Adapter X540-T1 (1x RJ-45)UPS: APC Smart-UPS 1000VA LCD 120V - 2U RackmountOperating System: Microsoft Windows Server 2012

, . , . , 32 GPU

34

Essential Hardware

Designers PC5CPUCore i7-4790 (3.6GHz)RAM32 GBHDD3 TBGPUNVIDIA GeForce GTX 760 (2GB)KeyboardGenius GK 110001MouseGigabyte GM-M6800Operating SystemWindows 8.1

Programmers PC2CPUCore i7-4790 (3.6GHz)RAM16 GBHDD2 TBGPUNVIDIA GeForce GTX 760 (2GB)KeyboardGenius GK 110001MouseGigabyte GM-M6800Operating SystemWindows 8.1

Essential HardwareOculus Rift (Augmented reality glasses)1Black Magic Cinema Camera1

, . , . , 32

36

Potential CustomersOil and Gas industriesMedical institutionsEducational and Research institutionsConstruction CompaniesAdministration of YekaterinburgPublic event organizersInformation technology companies.

Current project stage

Our offer for an investorOur current requirements 1-st investment round 350 000$Development of the cloud scientific utilities;Development and sale of software solutions for the Russian Industry and scientce;Creation of a Data-Center for scientific computing;Deployment of the cloud server applications ;2-nd Investment Round 1 100 000$.Development of the digital scientific environments for scienceBusiness scaling and improvement;