International Business Machines Corporation

Hewlett-packard Development Company, L.p.

Sequent Computer Systems, Inc.

Corene Casper

Brent A Kingsbury

Phillip E Krueger

Paul Edward McKenney

Terence M Rokop

Kenneth F Dove

Randal Craig Swanberg

Samuel Revitch

Michael J Callahan

Michael J Spitzer

Kenneth Frank Dove

Phil E Krueger

Phil Krueger

One embodiment is a method that uses a hierarchical bitmap tree in a storage system to store where available and unavailable memory space exists. Bitmap blocks are retrieved from the hierarchical bitmap tree to locate the available memory space.

Tracking memory space in a storage system

A storage area network is managed. In one embodiment, the method comprises providing a plurality of nodes; providing a plurality of storage, wherein the plurality of storage is shared by the plurality of nodes; determining if a change in the storage area network has occurred; and dynamically adjusting to the change if the change has occurred. In another embodiment, the system comprises a processor configured to communicate with a second node and at least one storage, wherein the storage is shared by the processor and the second node; the processor also being configured to determine if a change in the storage area network has occurred; and dynamically adjusting to the change if the change has occurred; and a memory coupled with the processor, the memory configured to provide instructions to the processor.

System and method for management of a storage area network

A system and method are disclosed for providing multi-node environment comprising a first node associated with a first operating system; a second node associated with a second operating system, wherein the second operating system is independent of the first operating system; a storage; and an interconnect coupling the first node with the storage and coupling the second node with the storage; and wherein the first node directly accesses the storage and the second node directly accesses the storage.

System and method for a multi-node environment with shared storage

An apparatus, method and computer program product for stopping processors in a multiprocessor system without using non-maskable interrupts are provided. With the apparatus, method and computer program product, at system initialization time, a copy of the operating system (OS) kernel is copied to a new physical location in memory. When a processor enters the debugger due to the occurrence of an event, the debugger switches its virtual-to-physical address mapping to point to the new copy of the OS kernel. The original copy of the OS kernel is then modified by inserting breakpoints, e.g., interrupts, in a repeating pattern in the text of the original copy of the OS kernel, with the exception of the breakpoint handler text in the original copy of the OS kernel. A cache flush of the remaining processors is then instigated thereby forcing the remaining processors to refetch instructions from the OS kernel. When the remaining processors fetch the OS kernel instructions, the instructions are fetched from the modified OS kernel. Thus, the processors encounter the inserted breakpoints and enter a breakpoint handler. The breakpoint handler then, by virtue of the switched virtual-to-physical address mapping, redirects the processor to the new copy of the OS kernel and handles the breakpoint in a normal fashion, e.g. causes the processor to enter the debugger.

Apparatus, method and computer program product for stopping processors without using non-maskable interrupts

A method for a user process to specify a policy for allocating pages of physical memory on the nodes of a multinode multiprocessor computer system. Through means such as a system call, an application program can specify to the operating system that physical pages of memory for an application-specified portion of virtual address space are to be physically allocated upon a specified set of nodes, subject to the additional selection criteria that the pages are to be allocated at first reference upon: 1) the node upon which the reference first occurs; 2) the node which has the most free memory, or 3) that the pages should be evenly distributed across the indicated set of nodes. In effect, the operating system remembers the specified allocation policy and node set from which the physical pages can be subsequently allocated as established by a system call. Subsequent use of the virtual address space for which the allocation policy is defined results in the memory being allocated accordingly. In this way, an application program can use memory with the memory-locality most advantageous to it. The preferred embodiments of the invention include extensions to the mmap and shmget functions of UNIX-based operating systems.

User specifiable allocation of memory for processes in a multiprocessor computer having a non-uniform memory architecture

An apparatus and method is disclosed for allowing a user process to advise or control an operating system to place a process on a particular node or a node within a set of nodes. When a user process makes a system call to the operating system to either create, execute, move, or attach a process, a specified system resource is passed to the operating system with the system call. This advises the operating system to store the process on a node containing the system resource identified in the system call. The user process can also require the operating system to store a process on a particular node.

Multiprocessor computer system with user specifiable process placement

A method for a user process to specify a policy for allocating pages of physical memory on the nodes of a multinode multiprocessor computer system. Through means such as a system call, an application program can specify to the operating system that physical pages of memory for an application-specified portion of virtual address space are to be physically allocated upon a specified set of nodes, subject to the additional selection criteria that the pages are to be allocated at first reference upon: 1) the node upon which the reference first occurs; 2) the node which has the most free memory, or 3) that the pages should be evenly distributed across the indicated set of nodes. In effect, the operating system remembers the specified allocation policy and node set from which the physical pages can be subsequently allocated, as established by a system call. Subsequent use of the virtual address space for which the allocation policy is defined results in the memory being allocated accordingly. In this way, an application program can use memory with the memory-locality most advantageous to it. The preferred embodiments of the invention include extensions to the mmap and shmget functions of UNIX-based operating systems.

For a multiprocessor computer having shared memory distributed across multiple nodes, a method and system for dynamically replicating data such as program text stored in memory on a first node to memory on a second node for use by a process executing on the second node. In response to a page fault generated by the process in accessing data, a determination is made whether the data is present in the memory of another node. If so, memory is allocated on the process's node for the data, and the needed data is copied from the other node to the process's node. The process's page table entry for the missing data is then modified to contain the physical address of the allocated memory, where the data is now stored. The method is implemented in a preferred embodiment by using novel data structures linked to the data structures that are typically created when a file is mapped to a process's virtual address space.

Growing community of inventors

Round Rock, TX, United States of America

Corene Casper