International Business Machines Corporation

Xiv Ltd.

Gobaru Limited

Yaron Revah

Ofir Zohar

Haim Helman

Dror Cohen

Shemer Schwartz

Efri Zeidner

Gal Zuckerman

Matan Arazi

Amihay Birenboim

Shemer Schwarz

Kariel Sendler

Shermer Schwartz

Systems and methods for implementing a decentralized consensus-based object-oriented platform with the scalability and bandwidth necessary to constitute a single unified infrastructure for web 3.0. Each object in the system represents a piece of data and/or an entity and/or a function and can be owned, controlled, and interacted with by its creator and/or by other objects. Messages are sent between the objects, validation clusters then reach a consensus among themselves per each of the messages received in conjunction with a respective one of the objects, and actions are finally made in conjunction with each of the messages by at least the validation clusters that are in consensus regarding the message, thereby assuring state-coherence among all objects. The platform has the capability/capacity to become 'the computer', which is a computational extension of “the internet”, with the potential to encompass virtually any activity, from social networks to financial systems, e-commerce and metaverse.

Consensus-based object-oriented platform

A redundant decentralized microservice architecture, in which each of at least selected some of the microservices is executed multiple times by multiple microservice computing nodes acting as mirror sites after reaching a distributed consensus regarding the correct way/order in which the microservices are to be executed. Clusters of redundant microservice computing nodes work in intra-cluster consensus when responding to remote procedure calls (RPCs) by activating the associated microservices multiple times, and then sending multiple RPCs to additional clusters of redundant microservice computing nodes. The process may repeat as a chain of inter-cluster microservice activation events that facilitate execution of multiple different microservices together constituting a single operation that is executed resiliently even under the most detrimental fault conditions, thus achieving fault tolerant computing that combines multiple mirror sites, multiple operators, consensus mechanisms and cryptography to achieve censorship resistance, decentralized operation and Byzantine Fault Tolerant computing utilizing microservices.

Resilient and decentralized microservice architecture

Disclosed is a data processing and/or storage system. The data processing and/or storage system includes at least two interfaces, wherein each of the at least two interfaces includes a non-dedicated communication port for communicating data to and form external data systems or clients based on a rule base.

System and method for mirroring data

One or more parameters of a received data service request may be compared against a lookup table. Based on one or more priority values stored in the lookup table and associated with the one or more request parameters, the data service request may be assigned a priority level.

System, method, and circuit for servicing a client data service request

Disclosed is a data processing and/or storage system. The data processing and/or storage system may include at least two interfaces, wherein each of the at least two interfaces includes a non-dedicated communication port for communicating data to and form external data systems or clients based on a rule base.

Various embodiments of the invention relate to an apparatus and a method of managing a snapshot storage pool (SSP) associated with a storage unit of a distributed data storage system. One apparatus includes a logic module and a processor. The logic module is adapted to provide a threshold corresponding to a ratio between a current amount of storage resources used for storing snapshots in the SSP and a total storage capacity defined for the SSP. The processor is adapted to trigger an action that may be effective for managing the SSP in response to the amount of storage resources used for storing snapshots in the SSP crossing the threshold.

Managing snapshot storage pools

A method for data distribution, including distributing logical addresses among an initial set of devices so as provide balanced access, and transferring the data to the devices in accordance with the logical addresses. If a device is added to the initial set, forming an extended set, the logical addresses are redistributed among the extended set so as to cause some logical addresses to be transferred from the devices in the initial set to the additional device. There is substantially no transfer of the logical addresses among the initial set. If a surplus device is removed from the initial set, forming a depleted set, the logical addresses of the surplus device are redistributed among the depleted set. There is substantially no transfer of the logical addresses among the depleted set. In both cases the balanced access is maintained.

Data allocation in a distributed storage system

Various embodiments of the invention relate to an apparatus and a method of managing a snapshot storage pool (SSP) associated with a storage unit of a distributed data storage system. According to some embodiments of the invention, the apparatus may include a logic module and a processor. The logic module may be adapted to provide a threshold corresponding to a ratio between a current amount of storage resources used for storing snapshots in the SSP and a total storage capacity defined for the SSP. The processor may be adapted to trigger an action that may be effective for managing the SSP in response to the amount of storage resources used for storing snapshots in the SSP crossing the threshold.

A method for data mirroring including receiving a sequence of transactions at a first storage node from a host indicating data to be stored by the node, storing the data therein, and delineating a first batch of the transactions. The node returns to the host an acknowledgment of having stored the data, delineates a second batch of the transactions received subsequent to the acknowledgment, and conveys the data to a second storage node. The first node also conveys to the second node a first record of the transactions in the first batch and a second record of the transactions in the second batch, and the data are stored in the second node responsively to the first and second records, so that all data indicated by the transactions in the first batch are stored by the second node prior to storing data indicated by the transactions in the second batch.

Parallel asynchronous order-preserving transaction processing

Some embodiments of the invention relate to an apparatus and a method of managing a snapshot storage pool (SSP) associated with a storage unit of a distributed data storage system. According to some embodiments of the invention, the apparatus may include a logic module and a controller. The logic module may be adapted to provide a threshold corresponding to a ratio between a current amount of storage resources used for storing snapshots in the SSP and a total storage capacity defined for the SSP. The controller may be adapted to trigger an action which may be effective for managing the SSP in response to the amount of storage resources used for storing snapshots in the SSP crossing the threshold.

Method, an apparatus and a system for managing a snapshot storage pool

A method for data distribution, including distributing logical addresses among an initial set of devices so as provide balanced access, and transferring the data to the devices in accordance with the logical addresses. If a device is added to the initial set, forming an extended set, the logical addresses are redistributed among the extended set so as to cause some logical addresses to be transferred from the devices in the initial set to the additional device. There is substantially no transfer of the logical addresses among the initial set. If a surplus device is removed from the initial set, forming a depleted set, the logical addresses oldie surplus device are redistributed among the depleted set. There is substantially no transfer of the logical addresses among the depleted set. In both cases the balanced access is maintained.

A system, method and circuit for efficiently managing a cache storage device. A cache storage device may include a cache management module. The cache management module may be adapted to generate a management unit and to associate the management unit with new data that is to be written into the cache. The cache management module may be further adapted to assign two or more allocation units for each management unit, to store the new data in the cache. A cache management module may include a management unit module. The management unit module may be adapted to generate management units associated with predefined global cache management functions. The cache management module may further include an allocation unit module in communication with the management unit module. The allocation unit module may be adapted to assign allocation units for storing data written into the cache.

Method, system and circuit for efficiently managing a cache storage device

Upon receiving a request for one or a set of data blocks associated with a given data segment, a disk cache controller may retrieve into cache the some part of the entire data segment from a disk. Each data segment on a disk may include a fixed number of data blocks, and all data segments may include the same number of data blocks. Data segments may be dynamically defined and their locations and sizes may vary from segment to segment. Data segments may be defined when data is written to the disk, or may be defined at a later point. A table associated with a cache controller may store information as to the physical location or address on a disk of the starting point and/or size of each data segment.

System, method, and circuit for retrieving data in data blocks into a cache memory from a mass data storage device based on a triggering event

A storage system, consisting of one or more data storage logical units (LUs) formed in physical media. The LUs are adapted to receive command and respond to the commands to store and recall data. The storage system further includes a plurality of ports, each port being adapted to maintain a respective LU command queue for each of the LUs, such that upon receiving a command directed to one of the LUs, the port places the received command in the respective LU command queue. The port converts the received command to one or more converted commands at least some of which are directed to the physical media of the one of the LUs. The port then conveys the commands directed to the physical media in an order determined by the respective LU command queue.

Growing community of inventors

Tel Aviv, Israel

Yaron Revah