Patents by Inventor Cheng-Yi Hsieh
Cheng-Yi Hsieh has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 11442925Abstract: The present technology proposes techniques for ensuring globally consistent transactions. This technology may allow distributed systems to ensure the causal order of read and write transactions across different partitions of a distributed database. By assigning causally generated timestamps to the transactions based on one or more globally coherent time services, the timestamps can be used to preserve and represent the causal order of the transactions in the distributed system. In this regard, certain transactions may wait for a period of time after choosing a timestamp in order to delay the start of any second transaction that might depend on it. The wait may ensure that the effects of the first transaction are not made visible until its timestamp is guaranteed to be in the past. This may ensure that a consistent snapshot of the distributed database can be determined for any past timestamp.Type: GrantFiled: June 29, 2018Date of Patent: September 13, 2022Assignee: Google LLCInventors: Wilson Cheng-Yi Hsieh, Alexander Lloyd, Peter Hochschild, Michael James Boyer Epstein, Sean Quinlan
-
Patent number: 11353917Abstract: The present technology proposes techniques for generating globally coherent timestamps. This technology may allow distributed systems to causally order transactions without incurring various types of communication delays inherent in explicit synchronization. By globally deploying a number of time masters that are based on various types of time references, the time masters may serve as primary time references. Through an interactive interface, the techniques may track, calculate and record data relative to each time master thus providing the distributed systems with causal timestamps.Type: GrantFiled: August 13, 2020Date of Patent: June 7, 2022Assignee: Google LLCInventors: Peter Hochschild, Alexander Lloyd, Wilson Cheng-Yi Hsieh, Robert Edman Felderman, Michael James Boyer Epstein
-
Publication number: 20220149170Abstract: A chip structure includes a substrate, a bottom conductive layer, a semiconductor layer, an interlayer dielectric layer, at least one electrode, and at least one top electrode. The substrate includes in order a core layer and a composite material. The bottom conductive layer is disposed on the bottom surface of the core layer, the semiconductor layer is disposed on the substrate, and an interlayer dielectric layer is disposed on the semiconductor layer. The at least one electrode is disposed between the semiconductor layer and the interlayer dielectric layer, and the at least one top electrode is disposed on the interlayer dielectric layer and electrically coupled to the at least one electrode.Type: ApplicationFiled: November 11, 2020Publication date: May 12, 2022Inventors: Hsiu-Mei Yu, Cheng-Yi Hsieh, Wei-Chan Chang, Chang-Sheng Lin, Chun-Yi Wu
-
Publication number: 20220012264Abstract: Paxos transactions are pipelined in a distributed database formed by a plurality of replica servers. A leader server is selected by consensus of the replicas, and receives a lock on leadership for an epoch. The leader gets Paxos log numbers for the current epoch, which are greater than the numbers allocated in previous epochs. The leader receives database write requests, and assigns a Paxos number to each request. The leader constructs a proposed transaction for each request, which includes the assigned Paxos number and incorporates the request. The leader transmits the proposed transactions to the replicas. Two or more write requests that access distinct objects in the database can proceed simultaneously. The leader commits a proposed transaction to the database after receiving a plurality of confirmations for the proposed transaction from the replicas. After all the Paxos numbers have been assigned, inter-epoch tasks are performed before beginning a subsequent epoch.Type: ApplicationFiled: September 27, 2021Publication date: January 13, 2022Inventors: Wilson Cheng-Yi Hsieh, Alexander Lloyd
-
Patent number: 11132381Abstract: Paxos transactions are pipelined in a distributed database formed by a plurality of replica servers. A leader server is selected by consensus of the replicas, and receives a lock on leadership for an epoch. The leader gets Paxos log numbers for the current epoch, which are greater than the numbers allocated in previous epochs. The leader receives database write requests, and assigns a Paxos number to each request. The leader constructs a proposed transaction for each request, which includes the assigned Paxos number and incorporates the request. The leader transmits the proposed transactions to the replicas. Two or more write requests that access distinct objects in the database can proceed simultaneously. The leader commits a proposed transaction to the database after receiving a plurality of confirmations for the proposed transaction from the replicas. After all the Paxos numbers have been assigned, inter-epoch tasks are performed before beginning a subsequent epoch.Type: GrantFiled: January 7, 2019Date of Patent: September 28, 2021Assignee: Google LLCInventors: Wilson Cheng-Yi Hsieh, Alexander Lloyd
-
Publication number: 20210185126Abstract: Throughput is preserved in a distributed system while maintaining concurrency by pushing a commit wait period to client commit paths and to future readers. As opposed to servers performing commit waits, the servers assign timestamps, which are used to ensure that causality is preserved. When a server executes a transaction that writes data to a distributed database, the server acquires a user-level lock, and assigns the transaction a timestamp equal to a current time plus an interval corresponding to bounds of uncertainty of clocks in the distributed system. After assigning the timestamp, the server releases the user-level lock. Any client devices, before performing a read of the written data, must wait until the assigned timestamp is in the past.Type: ApplicationFiled: March 2, 2021Publication date: June 17, 2021Inventors: Wilson Cheng-Yi Hsieh, Peter Hochschild
-
Patent number: 10951706Abstract: Throughput is preserved in a distributed system while maintaining concurrency by pushing a commit wait period to client commit paths and to future readers. As opposed to servers performing commit waits, the servers assign timestamps, which are used to ensure that causality is preserved. When a server executes a transaction that writes data to a distributed database, the server acquires a user-level lock, and assigns the transaction a timestamp equal to a current time plus an interval corresponding to bounds of uncertainty of clocks in the distributed system. After assigning the timestamp, the server releases the user-level lock. Any client devices, before performing a read of the written data, must wait until the assigned timestamp is in the past.Type: GrantFiled: December 9, 2016Date of Patent: March 16, 2021Assignee: Google LLCInventors: Wilson Cheng-Yi Hsieh, Peter Hochschild
-
Patent number: 10824438Abstract: A radio node device executes a backhaul connection method. The radio node device receives a radio access network issued configuration message requesting multi-connectivity capability of the relay node device. The radio node device provides two wireless communication channels in parallel as a part of a wireless backhaul channel to a radio access network entity in response to the configuration message. The relay node device serves as an intermediate node in the wireless backhaul channel. The relay node device performs route selection for the wireless backhaul channel based on metrics of relay nodes.Type: GrantFiled: February 25, 2019Date of Patent: November 3, 2020Assignee: CLOUD NETWORK TECHNOLOGY SINGAPORE PTE. LTD.Inventors: Chi-Chang Lu, Cheng-Yi Hsieh
-
Patent number: 10783133Abstract: The subject matter described herein provides techniques to ensure that queries of a distributed database observe a consistent read of the database without locking or logging. In this regard, next-write timestamps uniquely identify a set of write transactions whose updates can be observed by reads. By publishing the next-write timestamps from within an extendable time lease and tracking a “safe timestamp,” the database queries can be executed without logging read operations or blocking future write transactions, and clients issuing the queries at the “safe timestamp” observe a consistent view of the database as it exists on or before that timestamp. Aspects of this disclosure also provide for extensions, done cheaply and without the need for logging, to the range of timestamps at which read transactions can be executed.Type: GrantFiled: June 23, 2017Date of Patent: September 22, 2020Assignee: Google LLCInventors: Wilson Cheng-Yi Hsieh, Eric Hugh Veach, Michael James Boyer Epstein, Alexander Lloyd
-
Patent number: 10775837Abstract: The present technology proposes techniques for generating globally coherent timestamps. This technology may allow distributed systems to causally order transactions without incurring various types of communication delays inherent in explicit synchronization. By globally deploying a number of time masters that are based on various types of time references, the time masters may serve as primary time references. Through an interactive interface, the techniques may track, calculate and record data relative to each time master thus providing the distributed systems with causal timestamps.Type: GrantFiled: January 23, 2018Date of Patent: September 15, 2020Assignee: Google LLCInventors: Peter Hochschild, Alexander Lloyd, Wilson Cheng-Yi Hsieh, Robert Edman Felderman, Michael James Boyer Epstein
-
Publication number: 20190266176Abstract: Paxos transactions are pipelined in a distributed database formed by a plurality of replica servers. A leader server is selected by consensus of the replicas, and receives a lock on leadership for an epoch. The leader gets Paxos log numbers for the current epoch, which are greater than the numbers allocated in previous epochs. The leader receives database write requests, and assigns a Paxos number to each request. The leader constructs a proposed transaction for each request, which includes the assigned Paxos number and incorporates the request. The leader transmits the proposed transactions to the replicas. Two or more write requests that access distinct objects in the database can proceed simultaneously. The leader commits a proposed transaction to the database after receiving a plurality of confirmations for the proposed transaction from the replicas. After all the Paxos numbers have been assigned, inter-epoch tasks are performed before beginning a subsequent epoch.Type: ApplicationFiled: January 7, 2019Publication date: August 29, 2019Inventors: Wilson Cheng-Yi Hsieh, Alexander Lloyd
-
Publication number: 20190187999Abstract: A radio node device executes a backhaul connection method. The radio node device receives a radio access network issued configuration message requesting multi-connectivity capability of the relay node device. The radio node device provides two wireless communication channels in parallel as a part of a wireless backhaul channel to a radio access network entity in response to the configuration message. The relay node device serves as an intermediate node in the wireless backhaul channel. The relay node device performs route selection for the wireless backhaul channel based on metrics of relay nodes.Type: ApplicationFiled: February 25, 2019Publication date: June 20, 2019Inventors: CHI-CHANG LU, Cheng-Yi Hsieh
-
Patent number: 10191960Abstract: Paxos transactions are pipelined in a distributed database formed by a plurality of replica servers. A leader server is selected by consensus of the replicas, and receives a lock on leadership for an epoch. The leader gets Paxos log numbers for the current epoch, which are greater than the numbers allocated in previous epochs. The leader receives database write requests, and assigns a Paxos number to each request. The leader constructs a proposed transaction for each request, which includes the assigned Paxos number and incorporates the request. The leader transmits the proposed transactions to the replicas. Two or more write requests that access distinct objects in the database can proceed simultaneously. The leader commits a proposed transaction to the database after receiving a plurality of confirmations for the proposed transaction from the replicas. After all the Paxos numbers have been assigned, inter-epoch tasks are performed before beginning a subsequent epoch.Type: GrantFiled: January 5, 2016Date of Patent: January 29, 2019Assignee: GOOGLE LLCInventors: Wilson Cheng-Yi Hsieh, Alexander Lloyd
-
Patent number: 10042881Abstract: The present technology proposes techniques for ensuring globally consistent transactions. This technology may allow distributed systems to ensure the causal order of read and write transactions across different partitions of a distributed database. By assigning causally generated timestamps to the transactions based on one or more globally coherent time services, the timestamps can be used to preserve and represent the causal order of the transactions in the distributed system. In this regard, certain transactions may wait for a period of time after choosing a timestamp in order to delay the start of any second transaction that might depend on it. The wait may ensure that the effects of the first transaction are not made visible until its timestamp is guaranteed to be in the past. This may ensure that a consistent snapshot of the distributed database can be determined for any past timestamp.Type: GrantFiled: November 22, 2016Date of Patent: August 7, 2018Assignee: Google LLCInventors: Wilson Cheng-Yi Hsieh, Alexander Lloyd, Peter Hochschild, Michael James Boyer Epstein, Sean Quinlan
-
Patent number: 10037346Abstract: The subject matter described herein provides techniques to ensure that queries of a distributed database observe a consistent read of the database without locking or logging. In this regard, next-write timestamps uniquely identify a set of write transactions whose updates can be observed by reads. By publishing the next-write timestamps from within an extendable time lease and tracking a “safe timestamp,” the database queries can be executed without logging read operations or blocking future write transactions, and clients issuing the queries at the “safe timestamp” observe a consistent view of the database as it exists on or before that timestamp. Aspects of this disclosure also provide for extensions, done cheaply and without the need for logging, to the range of timestamps at which read transactions can be executed.Type: GrantFiled: October 26, 2012Date of Patent: July 31, 2018Assignee: Google LLCInventors: Wilson Cheng-Yi Hsieh, Eric Hugh Veach, Michael James Boyer Epstein, Alexander Lloyd
-
Publication number: 20180167460Abstract: Throughput is preserved in a distributed system while maintaining concurrency by pushing a commit wait period to client commit paths and to future readers. As opposed to servers performing commit waits, the servers assign timestamps, which are used to ensure that causality is preserved. When a server executes a transaction that writes data to a distributed database, the server acquires a user-level lock, and assigns the transaction a timestamp equal to a current time plus an interval corresponding to bounds of uncertainty of clocks in the distributed system. After assigning the timestamp, the server releases the user-level lock. Any client devices, before performing a read of the written data, must wait until the assigned timestamp is in the past.Type: ApplicationFiled: December 9, 2016Publication date: June 14, 2018Inventors: Wilson Cheng-Yi Hsieh, Peter Hochschild
-
Patent number: 9915970Abstract: The present technology proposes techniques for generating globally coherent timestamps. This technology may allow distributed systems to causally order transactions without incurring various types of communication delays inherent in explicit synchronization. By globally deploying a number of time masters that are based on various types of time references, the time masters may serve as primary time references. Through an interactive interface, the techniques may track, calculate and record data relative to each time master thus providing the distributed systems with causal timestamps.Type: GrantFiled: October 28, 2015Date of Patent: March 13, 2018Assignee: Google LLCInventors: Peter Hochschild, Alexander Lloyd, Wilson Cheng-Yi Hsieh, Robert Edman Felderman, Michael James Boyer Epstein
-
Patent number: 9786379Abstract: A data storage device including a flash memory and a controller. The flash memory includes a plurality of pages and a plurality of word lines, wherein each of the word lines controls at least two of the pages. The controller reads a first page of the pages in response to a read command, wherein the first page is controlled by a first word line of the word lines, and the controller further writes dummy data into the pages controlled by the first word line other than the first page when a predetermined condition is satisfied, wherein the predetermined condition includes that the first word line is not close.Type: GrantFiled: February 22, 2017Date of Patent: October 10, 2017Assignee: Silicon Motion, Inc.Inventors: Cheng-Yi Hsieh, Ming-Yen Lin
-
Patent number: 9774676Abstract: A system, computer-readable storage medium storing at least one program, and a computer-implemented method for identifying a storage group in a distributed storage system into which data is to be stored is presented. A data structure including information relating to storage groups in a distributed storage system is maintained, where a respective entry in the data structure for a respective storage group includes placement metrics for the respective storage group. A request to identify a storage group into which data is to be stored is received from a computer system. The data structure is used to determine an identifier for a storage group whose placement metrics satisfy a selection criterion. The identifier for the storage group whose placement metrics satisfy the selection criterion is returned to the computer system.Type: GrantFiled: May 21, 2013Date of Patent: September 26, 2017Assignee: GOOGLE INC.Inventors: Jeffrey Adgate Dean, Sanjay Ghemawat, Yasushi Saito, Andrew Fikes, Christopher Jorgen Taylor, Sean Quinlan, Michal Piotr Szymaniak, Sebastian Kanthak, Wilson Cheng-Yi Hsieh, Alexander Lloyd, Michael James Boyer Epstein
-
Patent number: 9747310Abstract: A method reads and writes data from a database table. Each row in the table has a primary key and multiple non-key columns. Each non-key column has one or more column values, and each column value has an associated timestamp that identifies when the column value was stored. The timestamps associated with the column values in each non-key column provide a unique order for the column values. A read transaction is initiated to read from a first non-key column of a first row. A write transaction is in progress that is updating a second non-key column of the first row, where the second non-key column is distinct from the first non-key column. The write transaction holds a lock on the second non-key column of the first row. The method concurrently reads the data from the first non-key column and writes a new column value to the second non-key column.Type: GrantFiled: June 4, 2013Date of Patent: August 29, 2017Assignee: GOOGLE INC.Inventors: Wilson Cheng-Yi Hsieh, Alexander Lloyd, Eric Hugh Veach