Eric Shannon Knauft

US Patent:

20220261386, Aug 18, 2022

Filed:

Feb 17, 2021

Appl. No.:

17/177686

Inventors:

- Palo Alto CA, US
Wenguang WANG - Santa Clara CA, US
Eric KNAUFT - Palo Alto CA, US

International Classification:

G06F 16/22
G06F 16/23
G06F 16/2455
G06F 16/28

Abstract:

A method for modifying key-value pairs of a B+ tree is provided. The method receives a request to modify a particular key-value pair. Each node of the tree has a modification number. The method traverses a path on the tree from the root node toward the particular node. The traversing includes upon reaching a parent node of the path, acquiring a shared lock on both the parent node and a child node one level below the parent node. Upon determining that the child node is the particular node, the method stores the modification number of the particular node, releases the shared lock on the particular node, compares a current modification number of the node with its stored number, and acquires an exclusive lock on the node if the numbers are the same. The method increments the current modification number of the node and modifies it while in the exclusive lock.

US Patent:

20220179710, Jun 9, 2022

Filed:

Dec 4, 2020

Appl. No.:

17/112618

Inventors:

- Palo Alto CA, US
Eric Shannon KNAUFT - San Francisco CA, US
Pascal RENAULD - Palo Alto CA, US
Yiqi XU - Mountain View CA, US

Assignee:

VMware, Inc. - Palo Alto CA

International Classification:

G06F 9/50
G06F 9/48

Abstract:

An efficient scheduling of IOs in a computing system using dynamic bandwidth regulation includes building up a shared regulator to limit the total IOPS scheduling among all IO classes at any given time. Reserved regulators may be used to place limits on the IOPS scheduled for a particular IO class at any given time. An outstanding IO window may also limit the overall number of outstanding IOs, and/or the bytes of outstanding IOs at any particular time. A first stage of IO scheduling may involve enforcing the reserved regulators to limit the IOPS scheduled for particular IO classes. A second stage of IO scheduling may involve enforcing the shared regulator to limit the total IOPS scheduled for all IO classes.

US Patent:

20230064693, Mar 2, 2023

Filed:

Aug 24, 2021

Appl. No.:

17/410673

Inventors:

- Palo Alto CA, US
Long YANG - Fremont CA, US
Maxime AUSTRUY - Pully, CH
Matthew B. AMDUR - Cambridge MA, US
Eric KNAUFT - San Francisco CA, US

International Classification:

G06F 12/0804
G06F 16/22
G06F 12/1009

Abstract:

The disclosure herein describes storing data using a capacity data storage tier and a smaller performance data storage tier. The capacity data storage tier includes capacity data storage hardware configured to store log-structured leaf pages (LLPs), and the performance data storage tier includes performance data storage hardware. A virtual address table (VAT) includes a set of virtual address entries referencing the LLPs. A tree-structured index includes index nodes referencing the set of virtual address entries of the VAT. Data to be stored is received, and at least a first portion of metadata associated with the received data is stored in the LLPs using the VAT, and at least a second portion of metadata associated with the received data is stored in the performance data storage tier. The architecture reduces space usage of the performance data storage tier.

US Patent:

20230055813, Feb 23, 2023

Filed:

Oct 20, 2022

Appl. No.:

18/048184

Inventors:

- Palo Alto CA, US
Enning XIANG - San Jose CA, US
Eric KNAUFT - Palo Alto CA, US
Pascal RENAULD - Palo Alto CA, US

International Classification:

G06F 3/06
G06F 9/48
G06F 9/38
G06F 9/50
G06F 9/52
G06F 9/455

Abstract:

The disclosure herein describes performing resynchronization (“resync”) jobs in a distributed storage system based on a parallelism policy. A resync job is obtained from a queue and input/output (I/O) resources that will be used during execution of the resync job are identified. Available bandwidth slots of each I/O resource of the identified I/O resources are determined. The parallelism policy is applied to the identified I/O resources and the available bandwidth slots. Based on the application of the parallelism policy, a bottleneck resource of the I/O resources is determined and a parallel I/O value is calculated based on the available bandwidth slots of the bottleneck resource, wherein the parallel I/O value indicates a quantity of I/O tasks that can be performed in parallel. The resync job is executed using the I/O resources, the execution of the resync job including performance of I/O tasks in parallel based on the parallel I/O value.

US Patent:

20230018790, Jan 19, 2023

Filed:

Jul 15, 2021

Appl. No.:

17/377128

Inventors:

- Palo Alto CA, US
Eric KNAUFT - San Francisco CA, US
Enning XIANG - San Jose CA, US
Ojan THORNYCROFT - Los Altos CA, US
Asit DESAI - Palo Alto CA, US
Varun SHAH - Fremont CA, US

International Classification:

G06F 11/20
G06F 3/06

Abstract:

The disclosure herein describes enhancing data durability of a base component of a data object using an unplanned delta component during transient fault unavailability. A base component of a data object becoming unavailable due to a transient fault is detected. A delta component associated with the base component is generated, wherein the delta component includes unwritten storage space with an address space and a tracking bitmap including a plurality of bits associated with data blocks of the address space of the delta component. The stale LSN with which the base component is associated is assigned to the delta component and the delta component is synchronized with an active component of the data object based on the assigned stale LSN. The delta component records write I/O targeted for the base component and, based on detecting the base component becoming available, the base component is synchronized with the delta component.

US Patent:

20230004525, Jan 5, 2023

Filed:

Jul 2, 2021

Appl. No.:

17/367233

Inventors:

- Palo Alto CA, US
Abhay Kumar JAIN - Cupertino CA, US
Wenguang WANG - Santa Clara CA, US
Eric KNAUFT - San Francisco CA, US
Enning XIANG - San Jose CA, US

International Classification:

G06F 16/11
G06F 16/14
G06F 12/10
G06F 3/06

Abstract:

An example method of upgrading a distributed storage object from a first version to a second version includes: querying metadata of a first component configured according to the first version of the distributed storage object, the metadata defining extents of data on a disk group of the first component; populating, for a second component configured according to the second version of the distributed storage object, logical and middle maps based on the metadata such that initial entries in the logical map point to initial entries in the middle map, and the initial entries in the middle map point to physical addresses of the disk group of the first component; and reading the data from the disk group of the first component and writing the data to a disk group of the second component while updating the initial entries in the middle map.

US Patent:

20210382825, Dec 9, 2021

Filed:

Jun 26, 2020

Appl. No.:

16/914149

Inventors:

- Palo Alto CA, US
Eric Knauft - Palo Alto CA, US
Vamsidhar Gunturu - Cupertino CA, US

Assignee:

VMware, Inc. - Palo Alto CA

International Classification:

G06F 12/10
H03M 13/15

Abstract:

Systems for accessing client data is described. A request to access a first data block is received. The request indicates a first logical address referencing the first data block. First mapping data is employed to identify a first physical addresses corresponding to the first logical addresses. The first mapping data encodes a first LOM transaction ID and candidate local addresses. The first mapping data is employed to identify the candidate local address and the first LOM transaction ID. A usage table is employed to determine the current status of the first LOM transaction ID. The candidate local address is employed to access the first data block. Second mapping data is employed to identify an updated local address of the set of local addresses. The updated local address currently references the first data block. The updated local address is employed to access the first data block.

US Patent:

20210382858, Dec 9, 2021

Filed:

Jun 5, 2020

Appl. No.:

16/894663

Inventors:

- Palo Alto CA, US
Vamsidhar GUNTURU - Cupertino CA, US
Eric KNAUFT - Palo Alto CA, US
Pascal RENAULD - Palo Alto CA, US

Assignee:

VMware, Inc. - Palo Alto CA

International Classification:

G06F 16/18
G06F 16/182
G06F 16/17
G06F 16/22
G06F 16/2457
G06F 12/02
G06F 12/0817

Abstract:

Techniques for efficiently storing client data blocks on a distributed-computing system are provided. The system includes a fast performance tier and a large capacity tier. The capacity tier stores the client data blocks in erasure encoded data stripes. The performance tier stores logical map data including an address map indicating a correspondence between logical addresses associated with a first layer of the system and physical addresses associated with a second layer. A method includes receiving a request to include additional client data blocks in the client blocks. The request indicates logical addresses for additional blocks. Corresponding physical addresses for additional block are determined. Each additional block is stored at the physical address. Additional logical map data is stored in the performance tier. Storing the additional logical map data includes updating the address map to indicate the correspondence between the logical addresses and the physical addresses for the additional blocks.