International Business Machines Corporation

Tamiya Onodera

Kiyokuni Kawachiya

Hiroshi Horii

Michiaki Tatsubori

Akira Koseki

Kazunori Ogata

Toyotaro Suzumura

Michihiro Horie

Hiroaki Nakamura

Hideaki Komatsu

Takeshi Ogasawara

Mikio Takeuchi

Miki M Enoki

Tatsuhiro Chiba

Derek Bruce Inglis

Akihiko Tozawa

Rei Odaira

Yohsuke Ozawa

Hiroshi Inoue

Vijay Sundaresan

Taiga Nakamura

Graeme Johnson

Trent Gray-Donald

Scott Ross Trent

Takesji Ogasawara

Mikio Takauchi

A method for reducing the cost of stack scanning in garbage collection (GC) includes, in the GC of the first-generation heap area, registering, in a nursery object reference list prepared for each thread, one or more addresses, within each stack, which each refer to a nursery object, and updating a scanning unnecessary area starting pointer such that the addresses listed in the nursery object reference list are included in the area from the bottom of the stack to the address pointed to by the scanning unnecessary area starting pointer. The method further includes, in the next GC of the first-generation heap area, for the area from the bottom of the stack to the address pointed to by the scanning unnecessary area starting pointer, performing the GC processing on the addresses included in the nursery object reference list.

Method, program, and system for reducing the cost of stack scanning

A computer-implemented method and a computer processing system are provided for increasing memory density in a memory using heap contraction. The method includes dividing the heap into a plurality of work regions including a last region and other regions such the last region is larger in size than the other regions. The method further includes calculating a size of the heap contraction. The method also includes forming a pair of the last region and one of the other regions that has a largest free portion. The method additionally includes executing intra-table compaction and inter-table compaction on the heap. The method further includes contracting the last region by subtracting a prescribed space from the last region.

Heap contraction for increasing memory density in cloud environment

Garbage collection methods and systems include adding a data object to one of multiple queues owned by a first garbage collection thread. The queues include a public queue and multiple private queues. A task is popped from one of the plurality of queues to perform garbage collection. The public queue is swapped with one of the private plurality of private queues if there are no tasks in the public queue.

Balanced double deques for eliminating memory fences in garbage collection

Garbage collection methods include adding a data object to one of multiple queues owned by a first garbage collection thread. The queues include a public queue and multiple private queues. A task is popped from one of the plurality of queues to perform garbage collection. The public queue is swapped with one of the private plurality of private queues if there are no tasks in the public queue.

A computer-implemented method and system are provided for reducing garbage collection overhead. The method includes specifying, by a hardware processor from a set of nodes of a Directed Acyclic Graph representing an application program, wherein each of the nodes represents a respective one of a plurality of computation tasks, any of the nodes which store a respective computation result for at least a threshold period of time as cache nodes. The method further includes allocating, by the hardware processor, the respective computation result of each of the cache nodes into a tenure area of a memory.

Reducing minor garbage collection overhead

A method of collecting data into a server from multiple client computers is provided as a first aspect of the invention. The method includes the steps of: reconstructing a collection network in a tree structure having the server acting as a highest node and the multiple client computers acting as child nodes every time the data is collected; the server broadcasting a parent-child declaration to nodes placed at the lower level; and each of the nodes transferring data to a node placed at the higher level.

Data collection method and system

A method for controlling priority of dynamic compilation by a computer system is disclosed. A task is executed by using a thread pooled in a thread pool. A metric related to the dynamic compilation is monitored. And then, determination is done whether the metric meets a predetermined criterion. In returning of the thread to the thread pool for next execution, a lowering of priority of the thread is caused if the metric is determined to meet the predetermined criterion. The lowering of priority of the thread may be caused by causing the thread to sleep for a period of time before the returning. The metric may be a length of a compilation queue for the dynamic compilation or a utilization rate of a compiler thread executing the dynamic compilation.

Controlling priority of dynamic compilation

A method and system are provided for reducing garbage collection overhead. The method includes representing, by a hardware processor, an application program by a Directed Acyclic Graph having a set of nodes. Each of the nodes represents a respective one of a plurality of computation tasks. The method further includes specifying, by the hardware processor from the set of nodes, any of the nodes which store a respective computation result for at least a threshold period of time as cache nodes. The method further includes allocating, by the hardware processor, the respective computation result of each of the cache nodes into a tenure area of a memory.

One or more embodiments are directed to creating a structure comprising a weak reference to an object, a strong reference to the object, a count of remote references to the object, and an identifier for the object, incrementing the count when a remote reference is created to the object and decrementing the count when a remote reference is removed, setting the strong reference in response to the count being greater than a first threshold in order to avoid a garbage collection of the object, clearing the strong reference in response to the count being less than a second threshold to enable a garbage collection of the object, and removing the structure and related resources based on a recognition of the object's collection through the weak reference.

Garbage collection of an object

A method and system for code generation and inlining. The method includes: scanning a code, where the code includes a method; determining whether the method includes a branch if the method is too large in size to be inlined, where the branch is based on runtime types of argument; profiling runtime types of actual arguments at a call site in the method if the method includes the branch; generating a polymorphic inline cache (PIC) code from a plurality of arguments, where the plurality of arguments are based on a result of the profiling step; generating a specialized method for a frequently appearing combination of the runtime types of argument and arguments that can be invoked from the PIC code; and inlining body of the specialized method into the call site if the specialized method has a body size within a permissible range.

Method and system for code generation and inlining

A system including multiple application servers for accessing shared data and a centralized control unit for centrally controlling a lock applied to the shared data by each of the application servers. Each application server includes a distributed control unit for controlling a lock applied to the shared data by the application server and a selection unit for selecting any one of distributed mode in which a lock is acquired from the distributed control unit or a centralized mode in which a lock is acquired from the centralized control unit.

System for accessing shared data using multiple application servers

A method, system, and program for recording an object allocation site. In the structure of an object, a pointer to a class of an object is replaced by a pointer to an allocation site descriptor which is unique to each object allocation site, a common allocation site descriptor is used for objects created at the same allocation site, and the class of the object is accessed through the allocation site descriptor.

Method and program for recording object allocation site

A computer implemented control method, article of manufacture, and computer implemented system for determining whether stack allocation is possible. The method includes: allocating an object created by a method frame to a stack. The allocation is performed in response to: calling a first and second instruction in the method frame; the first instruction causes an escape of the object, and the second instruction cancels the escape of the object; the object does not escape to a thread other than a thread to which the object has escaped, at the point in time when the escape is cancelled; the first instruction has been called before the second instruction is called; and the object does not escape in accordance with an instruction other than the first instruction in the method frame, regardless of whether the object escapes in accordance with the first instruction.

Growing community of inventors

Tokyo, Japan

Tamiya Onodera