International Business Machines Corporation

Other

Takuya Nakaike

Hideaki Komatsu

Tatsushi Inagaki

Takeshi Ogasawara

Moriyoshi Ohara

Maged M Michael

Toshihiko Koju

Rei Odaira

Motohiro Kawahito

Mari Abe Fukuda

Takanori Ueda

Toshio Suganuma

Ai Yoshino

Harold Wade Cain, Iii

Angel Nunez Mencias

Ravi Nair

Hiroshi Inoue

Hong Min

Petr Novotny

Shyam Ramji

Peng Wu

Qi Zhang

Takashi Ito

Venkatraman Ramakrishna

Ali Ijaz Sheikh

Hironobu Takagi

Lei Yu

Fumihiko Kitayama

Masahiro Hori

Yohei Ueda

Mari Abe

Shinichi Hirose

Teruo Koyanagi

Kohichi Ono

Kentaro Fukuda

Christopher S Smith

Goh Kondoh

Chun Hui Suen

Yu Chin Fabian Lim

Jose G Castanos

Hideaki Komacou

Soiichi Kawano

Peter James De Meo

An example operation may include one or more of storing one or more public keys of one or more policy services, receiving, via a signing service, a blockchain transaction that has content that has been verified by a policy service and includes a signature created by a private key of the policy service, verifying, via a hardware security module (HSM), the signature of the policy service based on a public key of the policy service from among the one or more public keys, and in response to a verification of the signature of the policy service, signing, via the HSM, the blockchain transaction using a wallet key of a user of the blockchain transaction.

Digital asset platform with HSM verification

A method creates a table of keys and values. Each key is an element of an input array which is an input of a machine-learning pre-processing pipeline, and each value is an output of the pipeline. The method measures (1) a hit rate H to the memo table, (2) an average time Tto look up the table, (3) an average time Tto execute the pipeline, and (4) a threshold Ton a number of elements of the input array. The method looks up the value in the table by using an element of the input array as a key when T×H>Tand the number of elements in the input array is less than T. The method calls the pipeline in place of the lookup for all of the remaining elements in the input array when the value is not in the table.

Memoizing machine-learning pre-processing and feature engineering

Techniques for refinement of data pipelines are provided. An original file of serialized objects is received, and an original pipeline comprising a plurality of transformations is identified based on the original file. A first computing cost is determined for a first transformation of the plurality of transformations. The first transformation is modified using a predefined optimization, and a second cost of the modified first transformation is determined. Upon determining that the second cost is lower than the first cost, the first transformation is replaced, in the original pipeline, with the optimized first transformation.

Streamlining data processing optimizations for machine learning workloads

A system and method for scheduling a business process including tasks, comprises a calculation unit, a determination unit, and a decision unit. The calculation unit is configured to calculate an estimated processing time required to execute the tasks. The determination unit is configured to calculate an estimated end time of a route including the tasks on the basis of the estimated processing time and schedule of a user to execute the tasks, and determine whether to apply speculative execution to the business process on the basis of the estimated end time. The decision unit is configured to decide to speculatively execute a task out of the tasks in the business process. The decision is made with reference to a remaining period for executing the task. The remaining period is calculated on the basis of a predicted execution timing of each task and a deadline of the business process.

Scheduling business process

A computer-implemented method is provided for program repository management. The method includes identifying commits in an upstream commit log of an upstream branch and commits in a development commit log of a development branch. The method further includes extracting the commits in the development commit log of the development branch. The method also includes identifying, by a hardware processor in the upstream commit log, a code which is identical or similar to the extracted commits from the commit log of the development branch. The method additionally includes showing the identified code as a commit candidate of change in an upstream program code.

Merging changes from upstream code to a branch

A computer-implemented scaling method, an apparatus, and a computer program product for detecting whether a performance of a system reaches a target performance are provided. The method includes scaling a layered software environment, including scaling a first layer software environment in the layered software environment in response to the performance of the system not reaching the target performance and scaling a second layer software environment that is above the first layer software environment in the layered software environment in response to the performance of the system not reaching the target performance despite the first layer software environment being scaled. The method also includes scaling hardware resources used for executing the layered software environment in the system in response to the performance of the system not reaching the target performance before scaling of the first layer software environment or after scaling of the second layer software environment.

Multi-layer autoscaling for a scale-up cloud server

A method is provided for building calling context trees of an application to identify a thread or a device affecting a throughput of a transaction relating to the application. The method obtains profiles of call stacks of multiple threads. The method obtains wake-up profiles which are the call stacks of a notifier thread and a waiter thread, from among the multiple threads, in a case that the waiter thread leaves from a sleep state. The method builds the calling context trees to indicate transitions of methods tracing callee relations and wake-up relations based on the profiles of the call stacks and the wake-up profiles, wherein each method is represented by a node having an average thread number. The method extracts and displays tree portions which are transitively reachable from a calling context tree for the transaction and which include the thread or device affecting throughput.

Identifying software and hardware bottlenecks

Described is a computer-implemented method of reordering condition checks. Two or more condition checks in computer code that may be reordered within the code are identified. It is determined that the execution frequency of a later one of the condition checks is satisfied at a greater frequency than a preceding one of the condition checks. It is determined that there is an absence of side effects in the two or more condition checks. The values of the condition checks are propagated and abstract interpretation is performed on the values that are propagated. It is determined that the condition checks are exclusive of each other, and the condition checks are reordered within the computer code.

Reordering condition checks within code

A computer-implemented method is provided for processing requests among microservices in a computer network. The method includes generating, by a processor, a method to return a body of a response from an Application Programming Interface (API) that aggregates the requests among the microservices to form aggregated requests. The method further includes generating, by the processor, another API that automatically receives the aggregated requests from the API by receiving an array of parameters of the original API and calling the method using each of the parameters of the array as a call parameter for the method. The method also includes optimizing, by the processor, a number of aggregated requests in a network layer of the computer network based on a wait time and a response time to receive and transfer requests between micro services.

Aggregating requests among microservices

Method of reordering condition checks between variables based on a frequency that check is satisfied

A method for executing a program region by a computer system with transactional memory support is disclosed. The computer system uses hierarchical locks for executing the program region. Determination is conducted whether a first condition related to a transaction abort is satisfied in beginning a transaction for the program region. If the first condition is satisfied, a bottom level lock corresponding to a bottom level resource among available resources is acquired to execute the program region in the transaction. If a second condition is determined to be satisfied, a next level lock corresponding to next level resource is acquired. If the acquired lock is a top level lock corresponding to a top level resource, the program region is executed without using the transaction.

Execution of program region with transactional memory

A system for parallel processing tasks by allocating the use of exclusive locks to process critical sections of a task. The system includes storing update information that is updated in response to acquisition and release of an exclusive lock. When processing a task which includes a critical section containing code affecting execution of the other task, an exclusive execution unit acquires an exclusive lock prior to processing the critical section. When the section has been processed successfully, the lock is released and update information updated. Meanwhile a second task, whose critical section does not contain code affecting execution of the other task may run in parallel, without acquiring an exclusive lock, via a nonexclusive execution unit. The nonexclusive execution unit determines that the second critical section has successfully completed if the update information has not changed during processing of the second critical section.

System, method, program, and code generation unit

Method of reordering condition checks

Growing community of inventors

Yokohama, Japan

Takuya Nakaike