Mechanism To Enhance PCIe Generation Switching
US-2024427710-A1 · Dec 26, 2024 · US
US9304951B1 · US · B1
| Field | Value |
|---|---|
| Publication number | US-9304951-B1 |
| Application number | US-201213449491-A |
| Country | US |
| Kind code | B1 |
| Filing date | Apr 18, 2012 |
| Priority date | Apr 18, 2011 |
| Publication date | Apr 5, 2016 |
| Grant date | Apr 5, 2016 |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
Disclosed is a policy-based I/O dispatcher that intercepts I/O requests from several individual volumes made to a shared storage system (such as a RAID array) and manages these I/O requests so that no single volume monopolizes the I/O bandwidth and excludes the other volumes from timely completing their own I/O requests. Also disclosed are a plurality of attributes that may be assigned to each volume sharing a disk array to effectively manage the sharing of the disk array among the volumes. The I/O policy manager may handle I/O requests from several sources including volume-based iSCSI I/O requests as well as the file-share-based XFS I/O requests.
Opening claim text (preview).
What is claimed is: 1. A method for managing I/O operations in a shared storage system, the method comprising: by a logical volume manager module, managing plurality of individual volumes backed by a RAID array comprising a RAID controller and a plurality of physical disks, wherein the logical volume manager is configured to process I/O requests sent from the plurality of individual volumes; by a Policy-Based I/O module which resides in the stack between the plurality of individual volumes and the logical volume manager, intercepting, requests sent from the plurality of individual volumes to the logical volume manager for processing on the RAID array; and by the Policy-Based I/O module, assigning a maximum throughput perform attribute and a target latency performance attribute to each volume of the plurality of individual volumes; by the Policy-Based I/O module, regulating processing of the I/O requests between each of the plurality of individual volumes and the to logical volume manager in accordance with the I/O attributed assigned to each volume, wherein regulating processing of the I/O requests comprises: the Policy-Based I/O dispatcher determining for each volume of the plurality of volumes whether a throughput of the volume exceeds the assigned maximum throughput performance attribute of the volume, and if the assigned throughput performance attribute of the volume is not exceeded, the Policy-Based I/O dispatcher then forwards intercepted I/O requests of the volume to the logical volume manager for processing on the RAID array; and if the assigned maximum throughput performance attribute of the volume is exceeded, the Policy-Based I/O dispatcher then determines whether a latency of the volume exceeds the assigned target latency performance attribute of the volume, if the assigned target latency of the volume is not exceeded, then, instead of forwarding intercepted I/O requests to the logical volume manager, the Policy-Based I/O dispatcher module caches intercepted I/O requests of the volume until the throughput of the volume longer exceeds the assigned maximum throughput performance attribute of the volume; and if the assigned target latency of the volume is exceeded and the assigned maximum throughput performance attribute of the volume is exceeded, then, instead of caching intercepted I/O requests, the Policy-Based I/O dispatcher forwards intercepted I/O requests of the volume to the volume manager for processing of the I/O on the RAID array while temporarily allowing the throughput of the volume to exceed the assigned maximum throughput performance attribute of the volume until the latency of the volume no longer exceeds the target latency performance attribute set for the volume. 2. The method of claim 1 , wherein the assigned throughput performance attribute includes at least one from the group comprising a maximum read data rate, a maximum write data rate, a maximum number of read operations per second, and a maximum number of write operations per second, and wherein the assigned latency performance attributes includes at least one from the group comprising a target average read latency and a target average write latency. 3. A storage system computer for managing I/O operations in a shared storage system, comprising: a processor; a memory communicatively connected to the processor; a logical volume manager module that manages a plurality of individual volumes backed by a RAID array comprising a RAID controller and a plurality of physical disks, wherein the logical volume manager is configured to process I/O requests sent from the plurality of individual volumes; and a Policy-Based I/O dispatcher module which resides in the stack between the plurality of individual volumes and the logical volume manager to intercept I/O requests sent from the plurality of individual volumes to the logical volume manager for processing on the RAID array, wherein the Policy-Based I/O dispatcher module is configured to: assign, to each volume of the plurality of individual volumes, a maximum throughput perform attribute and a target latency performance attribute; and regulate processing of the I/O requests between each of the plurality of individual volumes and the logical volume manager in accordance with the I/O attributed assigned to each volume; wherein regulating processing of the I/O requests comprises: the Policy-Based I/O dispatcher determining for each volume of the plurality of volumes whether a throughput of the volume exceeds the assigned maximum throughput performance attribute of the volume, and if the assigned throughput performance attribute of the volume is not exceeded, the Policy-Based I/O dispatcher then forwards intercepted I/O requests of the volume to the logical volume manager for processing on the RAID array; and if the assigned maximum throughput performance attribute of the volume is exceeded, the Policy-Based I/O dispatcher then determines whether a latency of the volume exceeds the assigned target latency performance attribute of the volume, if the assigned target latency of the volume is not exceeded, then, instead of forwarding intercepted I/O requests to the logical volume manager, the Policy-Based I/O dispatcher module caches intercepted I/O requests of the volume until the throughput of the volume no toner exceeds the assigned maximum throughput performance attribute of the volume; and if the assigned target latency of the volume is exceeded and the assigned maximum throughput performance attribute of the volume is exceeded, then, instead of caching intercepted I/O requests, the Policy-Based I/O dispatcher forwards intercepted I/O requests of the volume to the volume manager module for processing of the I/O on the RAID array while temporarily allowing the throughput of the volume to exceed the assigned maximum throughput performance attribute set of the volume until the latency of the volume no longer exceeds the assigned target latency performance attribute of the volume. 4. The storage system computer of claim 3 , wherein the assigned throughput performance attribute includes at least one from the group comprising a maximum read data rate, a maximum write data rate, a maximum number of read operations per second, and a maximum number of write operations per second, and wherein the assigned latency performance attributes includes at least one from the group comprising a target average read latency and a target average write latency.
for load management (allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029) · CPC title
Improving I/O performance · CPC title
for I/O devices · CPC title
by changing the path, e.g. traffic rerouting, path reconfiguration · CPC title
with latency improvement · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.