Abstract: A computing system can include memory management capabilities. In one embodiment, the system receives a request to allocate a portion of memory, where the request to allocate the portion of memory comprises a memory pool instance; determines a memory pool subinstance from which to allocate the portion of memory based on the memory pool instance; and obtains the portion of memory from the determined memory pool subinstance.

Abstract: A computing system can include memory management capabilities. In one embodiment, the system receives a request to update a first size of each of a plurality of portions of memory to a second size. The plurality of portions of memory can be associated with a first memory pool and the first memory pool can be associated with a memory pool instance. The system creates a memory pool subinstance based on the request. The memory pool subinstance can be associated with the memory pool instance. The system associates a second memory pool with the memory pool subinstance. The second memory pool comprises portions of memory of the second size and the second memory pool is associated with the memory pool instance.

Abstract: A network characteristic of current resources providing a network service may be identified. A determination as to whether the network characteristic of the current resources satisfies a load policy may be made. In response to the network characteristic of the current resources not satisfying the load policy, a new resource may be provisioned to provide the network service a first location associated with the new resource may be transmitted to a server that includes records associated with the network service. Otherwise, when the network characteristic of the current resources satisfies the load policy, an instruction to use an existing record at the server may be transmitted to the server.

Abstract: A computing system can include memory management capabilities. In one embodiment, the system receives a request to update a first size of each of a plurality of portions of memory to a second size. The plurality of portions of memory can be associated with a first memory pool and the first memory pool can be associated with a memory pool instance. The system creates a memory pool subinstance based on the request. The memory pool subinstance can be associated with the memory pool instance. The system associates a second memory pool with the memory pool subinstance. The second memory pool comprises portions of memory of the second size and the second memory pool is associated with the memory pool instance.

Abstract: A network characteristic of current resources providing a network service may be identified. A determination as to whether the network characteristic of the current resources satisfies a load policy may be made. In response to the network characteristic of the current resources not satisfying the load policy, a new resource may be provisioned to provide the network service a first location associated with the new resource may be transmitted to a server that includes records associated with the network service. Otherwise, when the network characteristic of the current resources satisfies the load policy, an instruction to use an existing record at the server may be transmitted to the server.

Abstract: A frame identifier running on a computing device generates identifier values and embeds the identifier values in a networking frame. The frame identifier also receives networking frames, and analyzes the received networking frames to identify a subset of received networking frames containing the identifier values. The identifier values indicate that corresponding received networking frames were originated from the network device. The frame identifier also discards the identified subset of received networking frames.

Abstract: A computing system can include memory management capabilities. In one embodiment, the system receives a request to update a first size of each of a plurality of portions of memory to a second size. The plurality of portions of memory can be associated with a first memory pool and the first memory pool can be associated with a memory pool instance. The system creates a memory pool subinstance based on the request. The memory pool subinstance can be associated with the memory pool instance. The system associates a second memory pool with the memory pool subinstance. The second memory pool comprises portions of memory of the second size and the second memory pool is associated with the memory pool instance.

Abstract: Provisioning a resource to provide a service based on network characteristics is disclosed. A request for a service may be received and network characteristics of current resources providing the service may be identified. If the network characteristics do not satisfy a load policy, a new resource is provisioned to provide the service. The location of the new resource may subsequently be transmitted. If the network characteristics do satisfy the load policy, then one of the current resources may be used to provide the service in response to the request.

Abstract: Systems and methods are disclosed for processing an input/output (I/O) operation. An example system includes a kernel interface that receives a notification of a page fault. The page fault is responsive to an application attempting to perform an operation on a memory region that is set to a first access mode. When the memory region is set to the first access mode, the application does not have permission to perform the operation on the memory region. The system also includes a handler that responsive to the notification (i) sets the memory region to a second access mode and (ii) spawns a kernel thread to drain data from the memory region. When the memory region is set to the second access mode, the application has permission to perform the operation on the memory region. The system further includes an I/O module that stores the data in the memory region for processing.

Abstract: Systems and methods are disclosed for processing an input/output (I/O) operation. An example system includes a kernel interface that receives a notification of a page fault. The page fault is responsive to an application attempting to perform an operation on a memory region that is set to a first access mode. When the memory region is set to the first access mode, the application does not have permission to perform the operation on the memory region. The system also includes a handler that responsive to the notification (i) sets the memory region to a second access mode and (ii) spawns a kernel thread to drain data from the memory region. When the memory region is set to the second access mode, the application has permission to perform the operation on the memory region. The system further includes an I/O module that stores the data in the memory region for processing.

Abstract: Methods and systems for centralized kernel module loading are described. In one embodiment, a computing system detects a kernel module load event to load a kernel module into a kernel of a client. Upon detection of the kernel module load event, the computing system computes a cryptographic hash of the kernel module, and sends the cryptographic hash to an access control server to verify whether the cryptographic hash is a permitted hash. The computing system receives a response from the access control server to permit or deny the kernel module load event, and permits or denies the kernel module load event based on the response.

Abstract: Provisioning a resource to provide a service based on network characteristics is disclosed. A request for a service may be received and network characteristics of current resources providing the service may be identified. If the network characteristics do not satisfy a load policy, a new resource is provisioned to provide the service. The location of the new resource may subsequently be transmitted. If the network characteristics do satisfy the load policy, then one of the current resources may be used to provide the service in response to the request.

Abstract: A method and system for network interface naming is described.

Abstract: A computing system can include memory management capabilities. In one embodiment, the system receives a request to update a first size of each of a plurality of portions of memory to a second size. The plurality of portions of memory can be associated with a first memory pool and the first memory pool can be associated with a memory pool instance. The system creates a memory pool subinstance based on the request. The memory pool subinstance can be associated with the memory pool instance. The system associates a second memory pool with the memory pool subinstance. The second memory pool comprises portions of memory of the second size and the second memory pool is associated with the memory pool instance.

Abstract: A frame identifier running on a computing device generates identifier values and embeds the identifier values in a networking frame. The frame identifier also receives networking frames, and analyzes the received networking frames to identify a subset of received networking frames containing the identifier values. The identifier values indicate that corresponding received networking frames were originated from the network device.

Abstract: A buffer manager allocates and logs network buffers for a non-uniform memory access (NUMA) machine is described. In one embodiment, the buffer manager receives information about an allocated network buffer. The buffer manager determines the allocation point of the buffer and logs information about the allocation point and a consumption profile information. This logged information is analyzed to determine a per-process buffer consumption profile. The NUMA machine is configured using the buffer consumption profile.

Abstract: Methods and systems for centralized kernel module loading are described. In one embodiment, a computing system detects a kernel module load event to load a kernel module into a kernel of a client. Upon detection of the kernel module load event, the computing system computes a cryptographic hash of the kernel module, and sends the cryptographic hash to an access control server to verify whether the cryptographic hash is a permitted hash. The computing system receives a response from the access control server to permit or deny the kernel module load event, and permits or denies the kernel module load event based on the response.

Abstract: A method and system for network interface naming is described.

Abstract: A buffer manager allocates and logs network buffers for a non-uniform memory access (NUMA) machine is described. In one embodiment, the buffer manager receives information about an allocated network buffer. The buffer manager determines the allocation point of the buffer and logs information about the allocation point and a consumption profile information. This logged information is analyzed to determine a per-process buffer consumption profile. The NUMA machine is configured using the buffer consumption profile.