BACKGROUND OF THE INVENTION

The invention relates generally to software applications and, more particularly, to methods and apparatus for managing resource usage in an object-based system.

In the consumer field, there is intense pressure to keep costs down. Thus, the capabilities of memory and CPU devices within consumer systems, for example, are scaled down as much as possible to minimize overall costs. For example, cable television set top boxes typically include limited memory space and CPU capabilities. As object-based systems are integrated within consumer devices, it is becoming more important to manage resource usage effectively within the object-based systems. In other words, as a result of limited resources (among other reasons), a major issue within the object-oriented field is how to track and manage resource usage.

In the JavaOS platform, the virtual machine is configured to classify resource availability into three states: RED, YELLOW and GREEN. Additionally, the virtual machine tracks resource availability, such as available Java heap space. If resource availability is within acceptable limits, resource availability is classified as being in GREEN mode. If a resource is reaching an unacceptable level of availability, resource availability is then classified as being in YELLOW mode. When in YELLOW mode, a request is sent to all applications to lower their resource usage. The notified applications are all expected to free up resources. If resource availability becomes worse (e.g., applications are not freeing up enough resources), the system can then go into a RED state. A RED state may indicate that one or more applications or other resource-consuming entities must be terminated to make resources available. As resources became available, the system can be put back into GREEN mode.

Although this conventional resource management scheme works well in some environments, it tends to be inefficient in systems in which multiple applications are simultaneously executing. When the multiple applications receive a request to lower their resource consumption, all of the applications may simultaneously begin freeing up resources. This conventional resource management schemes does not differentiate between applications. That is, all applications receive a same request to lower resource usage. Thus, a lower priority application may not necessarily lower its resource usage prior to a higher priority application lowering its resource usage. Additionally, an application that is using a disproportionate amount of resources may not relinquish its resources before several other applications that are minimally using resources relinquish all of the their resource usage and self-terminate. These scenarios are inefficient at best. Additionally, the virtual machine does not know the specifics of individual resource usage by each application.

US 5,748,468 describes a co-processor resource manager. The co-processor resource manager allocates resources of a co-processor to nodes implementing operations requested by applications running on a host processor. The resource manager functions in accordance with a user-modifiable prioritisation of the operations. Lower priority nodes can be evicted by the resource manager to resolve resource conflicts for newly requested operations. That is, before a new node is loaded, the resource manager determines whether loading the new node would create a resource conflict and takes appropriate action based upon that determination.

Accordingly, there is a need for improved mechanisms for efficiently managing resource usage by applications based on each individual application's resource usage characteristics or any other suitable criteria.

EP-A 1083485 discloses a method for managing resource usage of a particular resource by a set of related code, such as code executed on behalf of a downloaded applet. A resource indicator is associated with the related code, and the resource indicator indicates an amount of resource usage of the particular resource by the related code. The resource indicator is updated when the related code increases or decreases its collective resource usage of the particular resource.

SUMMARY OF THE INVENTION

The invention is defined in the appended claims.

Broadly speaking, the present approach fills these needs by providing apparatus and methods for managing resources. In general terms, a resource manager manages resource consumption of several resource entities which are each capable of consuming resources. The resource manager tracks the availability of such resources and determines whether a resource is critically short or reaches a particular usage level. When a resource becomes critically short or reaches a particular usage level, the resource manager selects one or more resource entities based on one or more criteria. For example, a resource entity which has the least restrictive resource usage policy or state is selected. The resource manager then requests that the selected resource entity changes its resource usage state to a more restrictive state. Of course, when resource usage reaches an acceptable level, the resource manager may also inform each resource entity (or previously selected resource entities) that they may set their resource consumption state to a less restrictive state.

Advantages of the present invention will become apparent upon reading the following detailed descriptions and studying the various figures of the drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention, together with further advantages thereof, may best be understood by reference to the following description taken in conjunction with the accompanying drawings in which:

• Figure 1 is a diagrammatic representation of a resource management system in accordance with one embodiment of the present invention;
• Figure 2 is a flowchart illustrating a procedure for managing resources in accordance with one embodiment of the present invention;
• Figure 3 is a flowchart illustrating the operation of Figure 2 for performing an episode with a selected resource entity (RE) to facilitate reclamation of resources from the selected RE in accordance with one embodiment of the present invention;
• Figure 4 is a diagrammatic representation of a resource management system with multiple and hierarchically arranged resource entities in accordance with an alternative embodiment of the present invention;
• Figure 5 is a flowchart illustrating a procedure for managing the resource consumption of multiple and hierarchical resource entities in accordance with an alternate embodiment;
• Figure 6 is a diagrammatic representation of a general-purpose computer system suitable for implementing the present invention; and
• Figure 7 is a diagrammatic representation of a virtual machine suitable for implementing the present invention.

DETAILED DESCRIPTION OF THE EMBODIMENTS

Reference will now be made in detail to specific embodiments of the invention. While the invention will be described in conjunction with specific embodiments, it will be understood that it is not intended to limit the invention to the described embodiments. On the contrary, it is intended to cover alternatives, modifications, and equivalents as may be included within the spirit and scope of the invention as defined by the appended claims. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. The present invention may be practiced without some or all of these specific details. In other instances, well known process operations have not been described in detail in order not to unnecessarily obscure the present invention.

In general, the invention pertains to methods and apparatus for a resource manager to individually manage multiple resource entities that each consume one or more resources, such as the Java heap. That is, the resource manager may request a selected resource entity (RE) to change its resource consumption state based on specific criteria. By way of example, the criteria for selecting an RE and/or the specific type of state change for the selected RE may include the current resource consumption state of each RE. In one case, the RE that is in a least restrictive resource consumption state (i.e., that will likely result in more resources being consumed by this RE than a more restricted RE) is selected and asked to change its resource consumption state to a more restrictive state (i.e., that will likely result in less resources being consumed by the selected RE). In another example, the criteria includes RE priority, and the RE that has the lowest priority is selected and requested to change its resource consumption state to a more restrictive state.

The resource consumption states that are possible for each RE may be defined or categorized in any suitable manner. In the illustrated embodiment, there are five states: NORMAL, FAVOR_SPACE, ESSENTIAL_ONLY, TERMINATE_GRACEFULLY, and TERMINATE_ABRUPTLY. An RE in a NORMAL state does not have to conserve resources. That is, the NORMAL state RE may use as much resources as it requires. An RE in a FAVOR_SPACE state favors space over speed. An RE in an ESSENTIAL_ONLY state conserves space by only using resources essential to proper operation. An RE in a TERMINATE_GRACEFULLY state gracefully shuts down its activities. In contrast, an RE in a TERMINATE_ABRUPTLY state immediately shuts down its activities. While some of these states may be communicated to a selected RE as a request for the RE to change its state, other states may simply be imposed on the selected RE by the resource manager. For example, the resource manager may cause the selected RE to be abruptly terminated.

Any suitable mechanism may be implemented for individually selecting and requesting one or more resource entity(s) to change their resource consumption state, and the selection and/or request may be based on any suitable criteria. Figure 1 is a diagrammatic representation of a resource management system 100 in accordance with one embodiment of the present invention. As shown, the system 100 includes a resource manager 102 configured to track the availability of resources 110. The resources may include any type of resource that is consumable by any type of software application or entity. By way of illustration, the resources 110 may include a Java heap, file descriptors, a native heap, video RAM, hardware resources such as MPEG Section filters, and persistent storage such as a file on a hard disk.. The resource manager 102 also manages one or more resource entities (e.g., 106 and 108) or "RE's" which each consumes one or more resource types, via a "ResourceNotificationList" 104. That is, resource manager 102 manages consumption of one or more resources 110 by one or more RE's.

Management of resources may be implemented through any suitable software and/or hardware construct. In the illustrated example, the resource manager 102 communicates resource consumption requests or changes to each RE through the ResourceNotificationList 104. Likewise, each RE communicates its own resource consumption policies or states to the resource manager 102 via ResourceNotificationList 104. In the illustrated embodiment, the ResourceNotificationList 104 includes specific callbacks for the resource manager to access and manage each resource entity. For example, the resource manager may request that a particular RE change its resource consumption state to a more restrictive state (i.e., which consumes less resources) via the ResourceNotificationList 104. In one response example, the particular RE may then respond via the ResourceNotificationList 104 that it has changed its current state to the requested state or request to remain in its current state.

Figure 2 is a flowchart illustrating a procedure 200 for managing resources in accordance with one embodiment of the present invention. The illustrated implementation of Figure 2 is generally a continuous process that includes setting up communication constructs between each new RE and the resource manager, monitoring computer resources, and making requests to selected RE's to change their resource usage based on such monitoring. Alternatively, resources may be periodically monitored in a process separate from the communication setup procedure. When a resource limit is reached, an activity may then be triggered that selects an RE and requests the selected RE to change its resource consumption state.

Referring back to the illustrated embodiment of Figure 2, one or more processes may be set up that may each consume one or more resources. For example, an application, such as a browser, may include one or more process entities, such as applets or Xlets. Each process entity may also be represented by one or more RE's that consume one or more resources. As shown, it is determined whether a new application is present in operation 201. If there is a new application, the new applications is set up in operations 202 through 208. Otherwise, these operations are skipped. In operation 202, the application is instantiated. The application may then instantiate one or more resource entity(s) (RE) in operation 204. The instantiated RE(s) are then each registered with the resource manager in operation 206.

The resource manager may then set a default state for each newly registered RE in operation 208. For instance, each RE is put in a NORMAL resource consumption state. Alternatively, each RE (or the controlling application) may set its own initial resource consumption state. Additionally, the resource manager may assume that each RE is set in a particular state (e.g., NORMAL) prior to making any requests for a state change to a selected RE.

The resource manager may then check for availability of resources in operation 210. It is then determined whether any resource is critically short in operation 212. A resource may be "critically short" when the available resource amount or percentage falls below a predetermined level or the resources usage amount or percentage rises above another predetermined level. If no resource is critically short, the resource manager continues to check the availability of resources in operation 210. If any resource is critically short, one or more RE are selected based on one or more criteria in operation 214. An RE may be selected based on a comparison of the relative resource consumption states of each RE or the relative priorities of each RE. For example, an RE having the lowest priority and/or the least restrictive resource consumption state (e.g., NORMAL) is selected. The resource manager then performs an episode with the selected one or more RE to reclaim resources in operation 216.

An episode is generally defined as a communication session between the resource manager and the selected RE. The episode includes the resource manager requesting the selected RE to change to a different state, the selected RE's response to such request, and/or the resource manager's response to the RE's response. The episode may end with the selected RE when the resource manager determines that the RE has satisfactorily complied with the request, the request no longer needs to be followed by the selected RE, or the RE is to be terminated.

Figure 3 is a flowchart illustrating the operation 216 of Figure 2 for performing an episode with a selected resource entity in accordance with one embodiment of the present invention. However, the illustrated process of Figure 3 represents merely one example of a resource manager's policy for managing resources with respect to a selected RE and is not meant to limit the scope of the invention to the described example policy. That is, the resource manager may implement any suitable resource management policy with respect to a selected RE.

As shown, the resource manager records the current time in operation 302. The recorded time is later used to determine how long the selected RE has spent responding to a resource shortage. The resource manager then sends a request to the selected RE to change its state to a more restricted state in operation 304. For example the resource manager requests that the selected RE change its NORMAL state to a FAVOR SPACE state. The resource manager determines what state to request of the selected RE in any suitable manner. In one example, the resource manager initially queries the RE for its current state. Alternatively, the resource manager may assume that the selected RE is in a particular default state (e.g., NORMAL) or in the last requested state.

It is then determined whether the selected RE is now in the requested state in operation 306. That is, it is determined whether the RE has satisfied the request from the resource manager. The resource manager may determine the state of the selected RE in any suitable manner. In one example, the resource manager may query the RE for its current state. In another embodiment, the RE responds to the request from the resource manager by indicating its current state. The system may also track resource usage of the entity (e.g., application) of which the RE is a part, instead of the RE changing its state.

If the RE is in the requested state, the resource manager again checks the availability of resources in operation 308. It is then determined whether any resource is still critically short in operation 310. If no resources are short, the episode ends in operation 314. Ending an episode may include informing the selected RE that it may change its state back to the less restrictive state (e.g., NORMAL).

However, if any resource is still critically short, it may then be determined whether the RE is already in the most restrictive state it can have in operation 312. If this is the most restrictive state for the RE, the episode ends in operation 314. Additionally, if resources are still critically short, the RE may be terminated. If this is not the most restrictive state for the RE, the resource manager then sends another request to the RE to change its state to a more restrictive state than its current state in operation 304 and the entire process repeats. For example, the RE is requested to go from a FAVOR_SPACE state to an ESSENTIAL_ONLY state.

If at some point the selected RE does not change to the request state, the resource manager then determines the time spent in this episode in operation 316. It may then be determined whether the elapsed time is greater than a specified limit in operation 318. For example, the limit may be set to one second. If the elapsed time is greater than the specified limit, it is then determined whether any resource is still critically short in operation 320. If no resource is short, the resource manager ends the episode with the RE without further action with respect to the selected RE. Alternatively, the resource manager may inform the selected RE that it may move to a less restrictive state (e.g., NORMAL). In an alternative approach, the RE is terminated as soon as the specified timeout limit is reached.

However, if a resource is still critically short, the resource manager terminates the unresponsive RE in operation 322 and then ends the episode in operation 314. If the elapsed time is not greater than the specified limit, it is then determined whether any resource is still critically short in operation 310. The operations for handling an RE that changed to a previously requested state are then followed. For example, the RE may again receive a request to change to a more restrictive state until resources are not critical.

The resource manager may alternatively determine whether to terminate an unresponsive RE based on whether the RE has indicated that it has reduced its resource consumption amount. For example, when an RE has reduced its consumption amount without changing its resource state, the resource manager may decide not to terminate the RE after the elapsed time has reached a first predetermined limit. In a further implementation, the RE may be terminated when the elapsed time reaches a second higher predetermined limit. In yet another embodiment, termination of the unresponsive RE may be based on whether another RE is available for sending a request to change its resource consumption state. That is, the unresponsive RE is not terminated when another RE is available for requesting a state change.

Any suitable termination technique may be implemented. One solution "thread.stop" is currently available as a resource within the Java™ II Standard Development Kit. Thread.stop throws an exception so that a currently executing method may "catch" the exception and execute any implemented clean up code. In one embodiment, the frames of a thread stack that are associated with the misbehaving code are popped from the thread stack. Exception handling code is allowed to execute for trusted code by popping the trusted code frame via processing an exception, but exception handling is not allowed to execute for untrusted code. In a second aspect, frames are popped on all thread stacks of all threads that are associated with the misbehaving code. Threads are generally deemed to be associated with the misbehaving code when the threads share a same object or the same resources as the thread that is associated with the misbehaving code.

In another embodiment, thread stacks associated with the misbehaving code are modified so that the thread aborts automatically after modification. A threads is generally deemed to be associated with the misbehaving code if the thread's stack has a program counter within a method associated with the misbehaving code (or the thread is executing methods associated with the misbehaving code). Frames within a stack that is associated with misbehaving code can be divided into two groups: trusted frames and untrusted frames. Trusted frames may be allowed to terminate on their own, while untrusted frames are modified to terminate without executing any remaining untrusted code. In other words, frames that reference untrusted code are not allowed to execute.

Some of the termination techniques described in the above referenced related patent applications include popping frames of a thread stack that are associated with misbehaving code (i.e., an unresponsive RE or an that is consuming excessive resources) are popped from the thread stack. Exception handling code is allowed to execute for trusted code by popping the trusted code frame via processing an exception, but exception handling is not allowed to execute for untrusted code. In a second aspect, frames are popped on all thread stacks of all threads that are associated with the misbehaving code. Threads are generally deemed to be associated with the misbehaving code when the threads share a same object or the same resources as the thread that is associated with the misbehaving code.

In other termination procedures, thread stacks associated with the misbehaving code are modified so that the thread aborts automatically after modification. A thread is generally deemed to be associated with the misbehaving code if the thread's stack has a program counter within a method associated with the misbehaving code (or the thread is executing methods associated with the misbehaving code). Frames within a stack that is associated with misbehaving code can be divided into two groups: trusted frames and untrusted frames. Trusted frames may be allowed to terminate on their own, while untrusted frames are modified to terminate without executing any remaining untrusted code. In other words, frames that reference untrusted code are not allowed to execute.

Figure 4 is a diagrammatic representation of a resource management system 400 with multiple and hierarchically arranged resource entities in accordance with an alternative embodiment of the present invention. As shown, the system 400 includes a first resource manager 402 that manages resources 410 and the resource usage of a first RE 406 and a second RE 408 through a first ResourceNotificationList 404. For example, the first RE 406 forms part of an application 412 such as a web browser and the second RE 408 is a video conference application. In this example, the application 412 also serves as a second resource manager that is configured to manage one or more of resources 410 and the resource usage of a third RE 416. By way of example, the third RE 416 is an applet or Xlet that consumes a cache that is managed by the browser application 412.

Figure 5 is a flowchart illustrating a procedure 500 for managing the resource consumption of multiple and hierarchical resource entities in accordance with an alternate embodiment. This procedure is described in conjunction with the system 400 of Figure 4. It is assumed that the ResourceNotificationList 404 is already set up for one or more RE's. Initially, it is determined whether the first ResourceNotificationList 404 has changed in operation 502. A change may include an addition of a registered new RE. If the first list 404 has changed, each newly registered RE (e.g., 406 or 408) associated with the first list 404 is then set to a default state in operation 504. If the first list 404 has not changed, operation 504 is skipped.

It is then determined whether any resources are critically short in operation 506. If no resource is critically short, the procedure 500 returns to the beginning operation 502. If a resource is critically short, an RE associated with the first list 404 that has the least restrictive state and/or lowest priority level is then selected in operation 508. It is then determined whether the selected RE is associated with an application that is a parent of another underlying RE in operation 510. In the illustrated example, the RE 406 is associated with an application 412 that is a parent of underlying RE 416.

If the selected RE is associated with a parent application, the first resource manager 402 performs an episode with the selected RE and the selected RE's associated application (e.g., 412) performs a second episode with one or more selected children RE in operation 512. In other words, the associated application serves as a resource manager of its underlying RE's. If the selected RE is not associated with a parent of another underlying RE, the first resource manager 402 performs an episode with only the selected RE in operation 514. The episodes may include any suitable resource management policies being implemented on a selected RE, for example, as described above with respect to Figure 3. The procedure 500 then repeats.

Figure 6 illustrates a typical, general-purpose computer system suitable for implementing the present invention. The computer system 1030 includes any number of processors 1032 (also referred to as central processing units, or CPUs) that are coupled to memory devices including primary storage devices 1034 (typically a read only memory, or ROM) and primary storage devices 1036 (typically a random access memory, or RAM). The computer system may take any suitable form. For example, the computer system may be integrated with a navigational system or television set top box.

Computer system 1030 or, more specifically, CPUs 1032, may be arranged to support a virtual machine, as will be appreciated by those skilled in the art. One example of a virtual machine that is supported on computer system 1030 will be described below with reference to Figure 7. As is well known in the art, ROM acts to transfer data and instructions uni-directionally to the CPUs 1032, while RAM is used typically to transfer data and instructions in a bi-directional manner. CPUs 1032 may generally include any number of processors. Both primary storage devices 1034, 1036 may include any suitable computer-readable media. A secondary storage medium 1038, which is typically a mass memory device, is also coupled bi-directionally to CPUs 1032 and provides additional data storage capacity. The mass memory device 1038 is a computer-readable medium that may be used to store programs including computer code, data, and the like. Typically, mass memory device 1038 is a storage medium such as a hard disk or a tape which generally slower than primary storage devices 1034, 1036. Mass memory storage device 938 may take the form of a magnetic or paper tape reader or some other well-known device. It will be appreciated that the information retained within the mass memory device 1038, may, in appropriate cases, be incorporated in standard fashion as part of RAM 1036 as virtual memory. A specific primary storage device - 1034 such as a CD-ROM may also pass data uni-directionally to the CPUs 1032.

CPUs 1032 are also coupled to one or more input/output devices 1040 that may include, but are not limited to, devices such as video monitors, track balls, mice, keyboards, microphones, touch-sensitive displays, transducer card readers, magnetic or paper tape readers, tablets, styluses, voice or handwriting recognizers, or other well-known input devices such as, of course, other computers. Finally, CPUs 1032 optionally may be coupled to a computer or telecommunications network, e.g., an Internet network or an intranet network, using a network connection as shown generally at 1012. With such a network connection, it is contemplated that the CPUs 1032 might receive information from the network, or might output information to the network in the course of performing the above-described mechanisms for managing resources. Such information, which is often represented as a sequence of instructions to be executed using CPUs 1032, may be received from and outputted to the network, for example, in the form of a computer data signal embodied in a carrier wave. The above-described devices and materials will be familiar to those of skill in the computer hardware and software arts.

As previously mentioned, a virtual machine may execute on computer system 1030. Figure 7 is a diagrammatic representation of a virtual machine which is supported by computer system 1030 of Figure 6, and is suitable for implementing the present invention. When a computer program, e.g., a computer program written in the Java™ programming language, is executed, source code 1110 is provided to a compiler 1120 within compile-time environment 1105. Compiler 1120 translates source code 1110 into byte codes 1130. In general, source code 1110 is translated into byte codes 1130 at the time source code 1110 is created by a software developer.

Byte codes 1130 may generally be reproduced, downloaded, or otherwise distributed through a network, e.g., network 1012 of Figure 6, or stored on a storage device such as primary storage 1034 of Figure 6. In the described embodiment, byte codes 1130 are platform independent. That is, byte codes 1130 may be executed on substantially any computer system that is running on a suitable virtual machine 1140.

Byte codes 1130 are provided to a runtime environment 1135 which includes virtual machine 1140. Runtime environment 1135 may generally be executed using a processor or processors such as CPUs 1032 of Figure 6. Virtual machine 1140 includes a compiler 1142, an interpreter 1144, and a runtime system 1146. Byte codes 1130 may be provided either to compiler 1142 or interpreter 1144.

When byte codes 1130 are provided to compiler 1142, methods contained in byte codes 1130 are compiled into machine instructions. In one embodiment, compiler 1142 is a just-in-time compile which delays the compilation of methods contained in byte codes 1130 until the methods are about to be executed. When byte codes 1130 are provided to interpreter 1144, byte codes 1130 are read into interpreter 1144 one byte code at a time. Interpreter 1144 then performs the operation defined by each byte code as each byte code is read into interpreter 1144. That is, interpreter 1144 "interprets" byte codes 1130, as will be appreciated by those skilled in the art. In general, interpreter 1144 processes byte codes 1130 and performs operations associated with byte codes 1130 substantially continuously.

When a method is invoked by another method, or is invoked from runtime environment 1135, if the method is interpreted, runtime system 1146 may obtain the method from runtime environment 1135 in the form of a sequence of byte codes 1130, which may be directly executed by interpreter 1144. If, on the other hand, the method which is invoked is a compiled method which has not been compiled, runtime system 1146 also obtains the method from runtime environment 1135 in the form of a sequence of byte codes 1130, then may go on to activate compiler 1142. Compiler 1142 then generates machine instructions from byte codes 1130, and the resulting machine-language instructions may be executed directly by CPUs 1032. In general, the machine-language instructions are discarded when virtual machine 1140 terminates. The operation of virtual machines or, more particularly, Java™ virtual machines, is described in more detail in The Java™ Virtual Machine Specification by Tim Lindholm and Frank Yellin (ISBN 0-201-63452-X).

Although only a few embodiments of the present invention have been described, it should be understood that the present invention may be embodied in many other specific forms. By way of example, the operations involved with managing resources may be reordered. Steps may also be removed or added.

Although the techniques for managing resources in accordance with the present invention are particularly suitable for implementation with respect to a Java™ based environment, the techniques may generally be applied in any suitable object-based environment. In particular, the techniques are suitable for use in platform-independent object-based environments. It should be appreciated that the methods may also be implemented in some distributed object-oriented systems.

By way of a further example implementation, the resource manager may manage resource usage according to the type of resource that is critically short. In this example, the selected resource entity may have a resource consumption state for each resource type and the resource manager requests that the selected resource entity changes its state for the specific type of resource which is critically short.

Accordingly, the present embodiments are to be considered as illustrative and not restrictive, and the invention is not to be limited to the details given herein, but may be modified within the scope and equivalents of the appended claims.