A static workflow scheduling method and device

A scheduling method and technology for scheduling devices, applied in program control devices, software simulation/interpretation/simulation, etc., can solve problems such as reduced performance of running instances, high economic costs, and failure to take into account the heterogeneity of resource use.

Active Publication Date: 2019-05-21
NORTHWEST UNIV
View PDF3 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, traditional cloud service providers do not take into account the heterogeneity of resource usage brought about by the increase of hardware devices, which greatly reduces the operating efficiency of workflows. For larger workflow structures, they must face One problem is processing in multi-cloud and multi-data centers. The hardware heterogeneity of data centers and the data transmission between data centers greatly limit the performance of the entire workflow. For high-throughput systems in the field of scientific computing , which limits the overall system performance
On the other hand, from the perspec

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A static workflow scheduling method and device
  • A static workflow scheduling method and device
  • A static workflow scheduling method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0066] Step 1: Workflow Layering

[0067] According to the specific application requirements and data placement submitted by users or cloud providers, a workflow structure G represented by a directed acyclic graph is generated w (V w ,E w), where Vw represents the set of modules in the workflow, and Ew represents the set of computing dependencies generated due to data placement and application requirements; the workflow is layered, there is no dependency relationship between modules in the same layer, and the total work of the first layer Quantity is CR l , with a budget of

[0068] Step 2: Sort by virtual machine type

[0069] For n (n greater than or equal to 1) types of virtual machines provided by multi-cloud providers, expressed as VT={vt 0 ,vt 1 ,...,vt n-1}, sort the virtual machine performance according to the priority of CPU, Memory, and Disk, that is, sort according to the size of CPU resources first, if the CPU resources are the same, sort according to the ...

Embodiment 2

[0090] For workflows and multi-cloud environments different from those in Example 1, based on the execution of the method in Example 1, there may be instances where the virtual machine type has been upgraded to the highest level during the virtual machine allocation process, and the total computing cost of the current level is still less than that of the current level A given budget, in this case, upgrade the virtual machine type to the next-level module of the current key module, and upgrade to the next level of the current virtual machine type, if the total computing cost of the current layer is less than the given budget of the current layer , then upgrade the virtual machine type to the second level for key modules, and compare the total computing cost of the current layer after the upgrade with the given budget of the current layer until the total computing cost of the current layer is greater than the given budget of the current layer.

Embodiment 3

[0092] For other different workflows and multi-cloud environments, on the basis of the solution in Embodiment 1 or 2, it may exist that in the process of selecting a physical machine, the total cost of the current layer exceeds the budget when using the worst bandwidth for transmission, and then reset the The virtual machine allocation plan is adjusted. Specifically, on the basis of the allocated virtual machine plan, the priority virtual machine type of the non-critical modules is downgraded. First, the virtual machine type of the lowest priority module among the non-critical modules is downgraded and the module priority is updated in time. , until the virtual machine type is reduced to the worst level or the budget requirement is met, continue to update and upgrade the bandwidth and compare.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a static workflow scheduling method and device. Hierarchical division is carried out according to the structure of the workflow, for the workflow module of each layer, the typeof the virtual machine is considered based on budget, the virtual machine is distributed, and then mapping scheduling of the physical machine is carried out on the distributed virtual machine. The performance bottleneck of the workflow in the heterogeneous multi-cloud environment is solved, and the defect of insufficient throughput of the workflow in the heterogeneous cloud environment is improved.

Description

technical field [0001] The invention belongs to the field of big data and cloud computing, in particular to a static workflow scheduling method and device in a multi-cloud environment. Background technique [0002] At present, applications in many natural science fields are generating more and more data at a steady rate. For the processing of these large amounts of data, workflow technology in cloud computing is often the basic solution. However, timely and effective solutions There are still challenges in analyzing and processing such large-scale data. Especially with the rapid growth of data dimension and scale, traditional high-performance computing (HPC) systems have difficulty meeting different levels of resource requirements due to their complex structures. [0003] With the emergence and maturity of cloud computing, computing resources can be flexibly allocated according to demand, and users only need to care about the amount of resources required by the correspondin...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06F9/455
Inventor 吴奇石许明睿侯爱琴严丽荣乔芮敏王永强
Owner NORTHWEST UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap