We Are Customer-Centric And Provide Customized Or One-Stop Full-Stack Solutions To Empower All Industries.
Bioinformatics (Sequence Search And Comparison Analysis)
The Calculation Is Integer Calculation, With Basically No Floating-Point Calculation. The Main Characteristics Of The Calculation Are Frequent Load And Write (Memory Reading And Writing), Which Means That A Larger Memory Capacity And Memory Bandwidth Are Required, And A Direct-Connect Architecture Cpu Is Required. The Network Pressure Is Not Great, And The Most Cost-Efective Gigabit Network Is Used, Without The Need To Use The More Expensive Infiniband High-Speed Network.
Molecular Dynamics
It Mainly Uses Floating-Point Calculations And Is Very Suitable For Large-Scale Parallelism, But It Has Large Network Requirements And Requires A Low-Latency, High-Bandwidth Infiniband High-Speed Network Between Nodes.
Molecular Docking (Drug Design)
The Amount Of Calculation Is Very Large And Consumes A Lot Of Computer Time; The Performance Requirements Of The Network Are Not High, And The Scalability Mainly Depends On The Number Of Ligand Small Molecules And The Uniformity Of Their Calculation Scale.
Sequencer Offline Processing
Compared With The Amount Of Computing, The Storage Capacity Requirement Is Greater; The Job Operation Needs To Be Combined With The Job Scheduling System; Offline Sequencing Processing Often Requires Sequence Splicing. Currently, The Most Mainstream Sequence Splicing Software, Denovo, Is A Multi-Threaded Program. The Single-Node Memory Capacity Requirement Is Very Large, Often Requiring 256g Or Even 512gb, And Requires The Configuration Of A Large-Capacity Memory Smp Fat Node.
Electron Microscope Image Processing
Mainly Floating-Point Calculations, Including A Large Number Of Single-Precision Fft Calculations; The Software Acceleration Ratio Is Completely Linear, And The Dependence On The Network Is Low; The Software Memory And Io Requirements Are Large, And Generally A Parallel File System Needs To Be Configured; The System Has A Large Amount Of Calculation, A Long Calculation Time, And High Requirements For System Stability And Reliability.
Mass Spectrometer Raw Data Processing
The Software Acceleration Ratio Is Close To Linear, And The Dependence On The Network Is Low; The System Has A Large Amount Of Calculation, A Long Calculation Time, And High Requirements For System Stability And Reliability.
Specifically, Gpus Are Designed To Solve Problems That Can Be Expressed As Data-Parallel Computations—Programs Executed In Parallel On Many Data Elements With Extremely High Computational Density (Ratio Of Mathematical Operations To Memory Operations). Since All Data Elements Execute The Same Program, There Is Less Demand For Precise Flow Control; And Because It Operates On Many Data Elements And Has A High Computational Density, Memory Access Latency Can Be Hidden Through Computation Without Having To Use Large Data Caches.
Data Parallel Processing Maps Data Elements To Parallel Processing Threads. Many Applications That Process Large Data Sets Can Use Data Parallel Programming Models To Accelerate Computation. In 3d Rendering, Large Sets Of Pixels And Vertices Are Mapped To Parallel Threads. Similarly, Image And Media Processing Applications (Such As Post-Processing Of Rendered Images, Video Encoding And Decoding, Image Scaling, Stereo Vision, And Pattern Recognition) Can Map Image Blocks And Pixels To Parallel Processing Threads. In Fact, Many Algorithms Outside The Field Of Image Rendering And Processing Are Also Accelerated By Data Parallel Processing – From General Signal Processing Or Physical Simulation To Mathematical Finance Or Mathematical Biology. In The Above Fields, Gpu Computing Has Been Successfully Applied And Achieved Incredible Acceleration Results.
The Boya Life Sciences High-performance Computing Cluster Is An Organic, High-performance, And Highly Reliable Cluster System. The System Hardware Uses Mainstream Products That Have Been Rigorously Tested To Ensure System Reliability; The Network That Interconnects The Various Components Of The System Is A Dedicated High-efficiency Network, And The Cluster Core Management System Enables The Entire System To Operate In A Coordinated Manner, Providing Users With Unified Services Like A Single High-performance Computer.
The Cluster Provides Users With A Single Computer Interface. The Front-end Computer Is Responsible For Interacting With Users And Assigning Tasks To Various Computing Nodes Through The Scheduler Program After Receiving The Computing Tasks Submitted By Users. After The Operation Is Completed, The Results Are Returned To The User Through The Front-end Computer. The Inter-process Communication During The Program Operation Is Carried Out Through A Dedicated Network.
The Solution Has Diversified Computing System Options. The Solution Is Equipped With 10 Pr8020p Dual-Core Eight-Core Blade Servers Developed By Powerleader. The Blade Cluster System Is Significantly Superior To Other Systems In Terms Of Computing Density, Power Consumption And Heat Dissipation, Operating Costs, Maintenance Costs, And Reliability, And Provides A Good Foundation For System Expansion And Upgrading.
The Computing System Can Also Use Powerleader’s High-performance Pr2510p Two-Way Server, Focusing On Floating Point Peak And Memory Performance. For Large Memory Requirements, Powerleader’s Self-developed Eight-way Pr8800g Smp Server Is Used, Using Intel Xeon E7 V4 Processor.
The Networks That Interconnect The Components Of The System Are All Dedicated And Efficient Networks, And The Entire System Runs In A Coordinated Manner Through The Cluster Core Management System.
It can provide high floating point computing performance for demanding life science calculations, and the computing cluster can flexibly adopt a variety of computing systems
The Computing Network Uses High-speed Edr 100gb Ib Network Interconnection To Ensure That The User's Computing Tasks Are Not Limited By Network Bandwidth During Operation, With Low Latency And High Speed. The Two-Layer High-speed Network Ensures That All Nodes In The Cluster Run Without Blocking And At Full Line Speed, Which Can Fully Meet The Needs Of High-Speed Interconnection.
Gpu Nodes, Storage Nodes, Computing Nodes, Etc. Have High Scalability, Which Can Not Only Meet The Current Business Requirements, But Also Be Upgraded And Expanded To Meet The Future Growth Of Business Volume.
The Entire Cluster Adopts A Variety Of Reliability Design Solutions To Greatly Improve The Reliability Of The System And Ensure That The System Has No Single Point Of Failure. At The Same Time, It Provides Users With A Simple And Easy-to-use Use And Maintenance Interface, Lowers The Threshold For System Use And Maintenance, And Improves The Maintenance Efficiency Of The Entire System.
