Hewlett-packard Development Company, L.P.

Juniper Networks, Inc.

William N Pohl

Bradd William Szonye

Suhas Suhas

Edward J Sharpe

Chukwuma Valentine Akpuokwe

Daniel DeFolo

Elizabeth An-Li Clark

Amit Ranpise

Robert Kentwortz

Walter J Searle

Alon Ronen

In general, techniques are described for monitoring memory usage in computing devices. A computing device comprising a memory and a control unit that executes a kernel of an operating system having kernel sub-systems may implement the techniques. A memory manager kernel subsystem, in response to requests for amounts of the memory from other one kernel sub-systems, allocates memory blocks from the memory. The memory manager, in response to requests to de-allocate one or more of the allocated memory blocks, determines the corresponding requested amounts of memory and sizes of the de-allocated blocks. The memory manager then generates memory usage information based on the determined requested amounts of memory and the determined ones of the two or more different sizes. The memory usage information specifies usage of the memory in terms of the two or more different sizes of the allocated plurality of memory blocks.

Monitoring memory usage in computing devices

It is determined that a memory pressure condition exists which limits how many active processes are allowed. There is generated and stored of a set of values corresponding to parameters for each process where the parameters are related to priority factors assigned to the associated process. There is calculated a prioritization score for each process based on the corresponding set of values. There is determined a first active process with the lowest priority based on the prioritization scores. The first active process is deactivated to reduce the memory pressure condition.

Prioritization of processes for deactivating processes to reduce memory pressure condition

In general, techniques of this disclosure relate to measuring scheduling performance of monitored threads in an operating system with improved precision. In one example, a method includes inserting, by an operating system kernel, a monitored thread into a queue comprising one or more threads and recording an insertion time that the monitored thread is inserted into the run queue; receiving, by the kernel, an event to remove the monitored thread from the run queue; responsive to receiving the event, determining, by the kernel, an amount of time that the monitored thread is stored on the run queue based on the insertion time and a removal time at which the monitored thread was removed from the run queue; and when the amount of time the monitored thread is stored on the run queue is greater than or equal to a specified threshold, sending a notification to a notification listener.

Measuring by the kernel the amount of time a monitored thread spends in a queue in order to monitor scheduler delays in a computing device

In an embodiment of the invention, an apparatus and method for a target thread selection in a multi-threaded process perform the steps of receiving a signal that may or may not be masked by threads in the process; and searching a thread subset for a target thread that can handle the signal. A signal daemon may search for the target thread if the target thread is not found in the thread subset.

Target thread selection in a multi-threaded process

An embodiment of the invention provides a concrete data type and a method for providing a cached chunked list concrete data type. The method can perform steps including: storing at least one datum in a chunk in a cache line; and setting a lower bit value (LB) in a link/space pointer in the chunk to indicate the empty slots in the chunk.

Cache chunked list concrete data type

A computer-implemented method of testing a target system using a stimulus-response test. The method includes obtaining the responses of the target system at predefined monitoring times. Each time the response changes state, a clock is reset, and the elapsed duration is kept from the latest response state change. If the elapsed duration from the last response state change exceeds an expected duration before the next state change, a possible fault is deemed to exist.

Growing community of inventors

Morgan Hill, CA, United States of America

William N Pohl