Sun Microsystems, Inc.

Yousef A Khalidi

Madhusudhan Talluri

Jose M Bernabeu-Auban

Vladimir Matena

Andrew G Tucker

Kenneth William Shirriff

Declan J Murphy

Moti N Thadani

Anil Swaroop

Michael N Nelson

Siamak Nazari

Charles E Narad

Srinivasan Viswanathan

Stephen Alan Chessin

Shing I Kong

Glen R Anderson

Vikram P Joshi

David Dion

Graham Hamilton

Panagiotis S Kougiouris

Dock G Williams

Basem A Nayfeh

A first computer sends a sequence of messages to a second computer using remote write operations to directly store each message in a corresponding memory location in the second computer. The second computer retains information denoting the sequence numbers of the messages it receives and processes, and it acknowledges each received message with an asynchronous acknowledgment message. The first computer keeps track of which messages it has sent but for which it has not yet received an acknowledgment. Whenever the first computer determines that it has failed to receive a message acknowledgment from the second computer in a timely fashion, or it needs to reuse previously used message sequence numbers, the first computer undertakes remedial actions to resynchronize the first and second computers. The process begins by prompting the second computer to flush and process all the messages in its receive FIFO, and then comparing sequence number information recorded by the second computer with the sequence numbers of the outstanding, unacknowledged messages sent by the first computer. If the comparison indicates that any messages sent by the first computer were not received and processed by the second computer, those messages are re-transmitted. If necessary, during resynchronization the first computer will activate a different communication interface than the one previously used so as to establish a reliable connection to the second computer. After a success resynchronization, normal &ldquo;send only&rdquo; message operation resumes. At predefined times, the sequence number information retained by the second computer is cleared.

System and method for message transmission between network nodes connected by parallel links

A method and apparatus are described that permit an application to control data transfer from a memory object of a source device to a sink device. The application can request that an operating system establish a mapping between a fast buffer and a memory object storing the data. The operating system then establishes the mapping between the fast buffer and the memory object thereby permitting the application to direct that the data of the memory object be transferred to the sink device. Thus, the sink device can use direct memory access to the source device to transfer the data from the memory object. Furthermore, if the application modifies a portion of the data of the memory object prior to directing the transfer, only the modified portion of the data is copied to main memory prior to transfer to the sink device.

Method and apparatus for an efficient data transfer mechanism

In a computer system having a kernel supporting an interface definition language, a file system composed of an object framework. The object framework includes a set of proxy vnodes, a set of memcache vnodes, and a set of storage vnodes. The set of proxy vnodes and the set of memcache vnodes are linked to the set of storage vnodes through the use of the interface definition language, and the set of proxy vnodes are linked to the set of memcache vnodes through the use of a set of pointers. Each proxy vnode of said set of proxy vnodes is typed so as to differentiate between a set of file system objects such as files, directories and devices. The set of memcache vnodes forms an interface to a virtual memory system while the set of storage vnodes forms an interface to an underlying file system. The file system also uses a set of file paging interfaces that support extensions to the file system while providing full coherence of data.

Method and apparatus for extending traditional operating systems file systems

One embodiment of the present invention provides a method and an apparatus that facilitates transparent failovers from a primary copy of an object on a first server to a secondary copy of the object on a second server when the first server fails, or otherwise becomes unresponsive. The method includes detecting the failure of the first server; selecting the second server; and reconfiguring the second server to act as a new primary server for the object. Additionally, the method includes transparently retrying uncompleted invocations to the object to the second server, without requiring explicit retry commands from a client application program. A variation on this embodiment further includes winding up active invocations to the object before reconfiguring the second server to act as the new primary server. This winding up process may include causing invocations to unresponsive nodes to unblock and complete. Another variation includes blocking new invocations to the object after detecting the failure of the first server, and unblocking these new invocations after reconfiguring the second server to act as the new primary server. Hence, the present invention can greatly simplify programming of client application programs for highly available systems. It also makes it possible to use a client application program written for a nonhighly available system in a highly available system.

Method and apparatus for transparent server failover for highly available objects

A file disaster recovery system that employs geographical replication of data from a local site to remote site in a such a manner that file requests from clients of the local site can be handled by a file server on the remote site following a failover from the local site to the remote site. Geographical data replication software running on a local server checkpoints to a log in local stable storage all information on file operations that change the file state of the local file system. According to a selected mode, the local geographical data replication software flushes information in the log pertaining to the file operations since the last flush to the remote site. At the remote site, compatible remote geographical data replication software running on a remote file server receives the flushed log and replicates in sequence order the file operations represented in the flushed log. The results of the operations are stored on remote stable storage. The local and remote servers can be clusters or single servers. There is no need for commonality, except for operating and file systems, between the local and remote sites. Because operations are replicated and not low level, formatted data, the local and remote stable file storage can be completely different.

Method and apparatus for file system disaster recovery

A system and method are disclosed that provides transparent, global access to devices on a computer cluster. The present system generates unique device type (dev.sub.-- t) values for all devices and corresponding links between a global file system and the dev.sub.-- t values. The file system is modified to take advantage of this framework so that, when a user requests that a particular device, identified by its logical name, be opened, an operating system kernel queries the file system to determine that device's dev.sub.-- t value and then queries the a device configuration system (DCS) for the location (node) and identification (local address) of a device with that dev.sub.-- t value. Once it has received the device's location and identification, the kernel issues an open request to the host node for the device identified by the DCS. File system components executing on the host node, which include a special file system (SpecFS), handle the open request by returning to the kernel a handle to a special file object that is associated with the desired device. The kernel then returns to the requesting user a file descriptor that is mapped to the handle, through which the user can access the device.

System and method for transparent, global access to physical devices on

The present invention pertains to a system and method for performing remote object invocation. A object-oriented computing system includes a number of independent computing nodes that are interconnected via a communications link. The nodes represent client and/or server computers that do not share memory. Each node includes a number of domains having separate address spaces. Each domain includes one or more threads of execution that invoke one or more objects. The object's method can reside in the same domain as the requesting thread, in a different domain within the same node, or in a different domain in another node. A file descriptor is used to represent those objects whose methods reside in a different domain than the requesting thread. A file descriptor is a protected kernel entity that enables a thread to invoke an object. A thread can only access those objects for which it has received an associated file descriptor. An Object Request Broker (ORB) services the object invocation requests for those objects residing in different domains. The ORB provides a mechanism that translates the file descriptor used to reference an object in one domain into the file descriptor used to invoke the object in the domain having the object's method.

System and method for remote object invocation

The present invention pertains to a system and method for tracking object references in an object-oriented computing system including a number of independent computing nodes interconnected by a communications link. The reference counting mechanism tracks references to an object through the use of a messaging protocol. A server node keeps a foreign reference count for each of its objects. The foreign reference count indicates the number of remote nodes having a reference to one of the servers objects. A server node increments the foreign reference count for each object reference that it exports prior to sending it to an intended client node. A client node will send a message to the server node when it has already received the exported object reference and in response to this message, the server node decrements the appropriate foreign reference count. When a client node receives an object reference from another client node, the receiving client node will message the appropriate server node that the receiving client node has a reference to the object. In response, the server node increments the appropriate foreign reference count. The reference counting mechanism utilizes a node failure protocol in order to accurately track the foreign reference counts in light of node failures.

Method and system for reliable remote object reference management

A system and method are disclosed for rendering devices on a cluster globally visible, wherein the cluster includes a plurality of nodes on which the devices are attached. The system establishes for each of the devices in the cluster at least one globally unique identifier enabling global access to the device. The system includes a device registrar that creates the identifiers and a global file system. The identifiers include a globally unique logical name by which users of the cluster identify the device and a globally unique physical name by which the global file system identifies the device. The registrar creates a one-to-one mapping between the logical name and the physical name for each of the devices. The system also includes a device information (dev.sub.-- info) data structure maintained by the device registrar that represents physical associations of the devices within the cluster. Each association corresponds to the physical name of a device file maintained by the global file system. The device registrar determines for an attached device a globally unique, device type (dev.sub.-- t) value; creates dev.sub.-- info data structure entry and a corresponding physical name; generates a logical name based on the dev.sub.-- t value and the physical name; and associates the dev.sub.-- t value with the device file representing the attached device. Given this framework, a user of the cluster can access any of the devices by issuing the global file system an access request identifying the device to be accessed by its logical name.

Global file system-based system and method for rendering devices on a

A method and apparatus for translating a virtual address to a physical address. A virtual address to be translated has a virtual page offset and a virtual page number. The virtual address to be translated addresses a page of memory. The size of this page is unknown. There are L different possible page sizes where L is a positive integer greater than one. Each of the L different page sizes is selected to be a test page size and a test is performed. During the test, a pointer into a translation storage buffer is calculated. The pointer is calculated from the virtual address to be translated by assuming that the virtual address to be translated corresponds to a mapping of the test page size. The pointer points to a candidate translation table entry of the translation storage buffer. The candidate translation table entry has a candidate tag and candidate data. The candidate tag identifies a particular virtual address and the candidate data identifies a particular physical address corresponding to the particular virtual address. A virtual address target tag is extracted from the virtual address to be translated. The virtual address target tag is calculated by assuming that the virtual address to be translated corresponds to a mapping of the test page size. The target tag and the candidate tag are then compared. If the target tag matches the candidate tag, the candidate data is provided as the physical address translation corresponding to the virtual address to be translated.

Virtual address to physical address translation of pages with unknown

A bit map is maintained by a provider object of a name server to keep track of names cached by a cache object of the client. The bit map is indexed by performing a hash of the name. When a name is looked up by the server on behalf of a client, the server hashes the name, and sets the bit in the bit map indexed by the result of the hash modulo the size of the bit map. The result of the hash is returned to the client and is stored with the entry in the cache. A bit 'set' in the bit map indicates that the client caches at least one name that hashes into the bit. When the server invalidates a name, a hash of the name to be invalidated is used to find the corresponding bit in the bit mask. If the bit is set, the server sends an invalidation request to the client. The invalidation request includes the result of the hash, and the size of the provider's bit map. The client invalidates all entries that hash into the specified bit in the bitmap on the server. Alternatively, the invalidation request also includes the name to be invalidated and the client invalidates only the entry that matches the name.

Memory efficient directory coherency maintenance

The present invention pertains to a system and method for tracking object references in an object-oriented computing system including a number of independent computing nodes interconnected by a communications link. The reference counting mechanism tracks references to an object at three levels. At a first level, the server handler associated with an object is used to track local references to the object by other user applications within the same domain. At a second level, the inter-process communications facility (i.e., door facility) is used to track object references that are exported to different domains within the same node. At a third level, the kernel object request broker (ORB) is used to track object references that are exported outside the node. When all external object references are released, the ORB notifies the object's server handler of this occurrence. Similarly, the door facility notifies the object's server handler when all inter-domain references have been released. The object's server handler will initiate the appropriate reclamation procedures when it receives notification from the ORB and the door facility and when its own local references have been relinquished.

System and method for remote object resource management

Transparent routing within the cluster is achieved (without changing the networking code on each node of the cluster) by using a pair of modules interposed appropriately on the networking stack. In a 'clustered' system built out of several computers, using the present invention, the networking subsystem appears to applications as if the applications are running on a single computer. In addition, no modifications to the networking code is needed. The present invention is extensible to a variety of networking protocols, not just TCP/IP as the packet filter allows the routing within the cluster to be done dynamically. No modifications to the applications is needed (same binaries will work). A packet filter and remote communication between the modules through IDL enable the modules to do their job. A name server that maintains the port name space is used.

Single-system image network subsystem in a clustered system

A computer system having virtual memory that can be mapped using multiple page sizes onto logically addressable physical memory. An intermediate addressing scheme permits the mapping of several non-contiguous small pages in physical memory onto a bigger sized virtual memory page. Rather than translating a virtual address directly into a physical address, a virtual address is translated into an intermediate address that may or may not be a physical address. If the virtual page is backed by physical memory that is contiguous and aligned on a proper boundary for the page size, then the intermediate address will be the physical address and no second translation is required. If the intermediate address is not a physical address, it is then translated into a physical address. This is the case where a big page in virtual memory is backed by more than one smaller page in physical memory. Thus, non-contiguous small pages in physical memory can be mapped together using an intermediate translation to form a single big page thereby removing the requirement that a big page be mapped using a single contiguous portion of physical memory and further removing the requirement that the big page be big page boundary aligned within physical memory. Furthermore, several small pages can be promoted to a single big page simply by changing the virtual address to intermediate address mappings and also changing the intermediate address to physical address mappings to reflect the promotion thereby eliminating the need to move the contents of the small pages into a single contiguous, big page aligned region of physical memory. Furthermore, a big page sized region of virtual memory that has one or more smaller page sized holes within it can be treated as a single big virtual memory page and be backed in physical memory using only as many smaller pages as are required to back the non-hole regions of the virtual address space.

Growing community of inventors

Sunnyvale, CA, United States of America

Yousef A Khalidi