Abstract: Systems and techniques are described for transferring data. A described technique includes receiving a request to transmit a data block from a first data storage device to a second data storage device. An attempt to read the data block from the first data storage device is made. A media error resulting from the attempt to read the data block from the first data storage device is detected. In response to detecting the media error, a new data block is generated and includes mismatched checksum data that causes a checksum mismatched error when the new data block is accessed. The new data block is transmitted for storage at the second data storage device in place of the data block.

Abstract: Systems and techniques are described for transferring data. A described technique includes determining that a first replica of an object stored at a first host has become available to a distributed storage system after previously being unavailable to the distributed storage system. The object includes a range of memory addresses at which data of the object is stored. In response to determining that the first replica has become available, resyncing data for the first replica is obtained. The resyncing data indicates whether each range of memory addresses is synchronized at the first replica with other replicas of the object. Tracking data for the first replica is obtained. The tracking data indicates whether data stored at the range of memory addresses of the object has been modified at a second replica while the first replica was unavailable. The resyncing data is updated based on the tracking data.

Abstract: Systems and techniques are described for transferring data. A described technique includes receiving a request to transmit a data block from a first data storage device to a second data storage device. An attempt to read the data block from the first data storage device is made. A media error resulting from the attempt to read the data block from the first data storage device is detected. In response to detecting the media error, a new data block is generated and includes mismatched checksum data that causes a checksum mismatched error when the new data block is accessed. The new data block is transmitted for storage at the second data storage device in place of the data block.

Abstract: A method of retransmitting a multicast message through a unicast channel in a virtual distributed storage system comprising a cluster of nodes that includes a master node and a set of agent nodes is provided. Each node maintains a copy of a directory of the virtual distributed storage system. The master node multicasts messages to the agent node through a multicast channel to update the directory. The method determines that a particular message to update the directory exceeds a maximum size limit for transmit to the agent nodes through the multicast channel. The method multicasts a retransmission trigger message from the master node to each agent node. The method receives a retransmission request message at the master node from each of the agent nodes through a unicast channel. The method sends the particular message to update the directory from the master node to each agent node through the unicast channel.

Abstract: An in-memory cache for a computer system having a first storage and a second storage where the first storage is a cache for the second storage, tracks priority levels of block attributes stored therein. If a data item is cached in the first storage, the block attribute corresponding to the data item is stored in the in-memory cache as a high priority block attribute. If a data item evicted from the first storage, the block attribute corresponding to the data item is stored in the in-memory cache as a low priority block attribute. When the cache becomes full, the low priority block attributes are evicted before the high priority block attributes.

Abstract: Techniques are disclosed for message retransmission. In one embodiment, an agent in a distributed system that receives and/or relays messages from a master node identifies messages that have not been successfully delivered, and adds a stub for each such message to a message queue. The agent then requests retransmission of missing messages in the message queue and sets flags associated with the message stubs to indicate that retransmission requests have been sent. If one (or more) of the messages cannot be retransmitted, the master node sends acknowledgment message(s) notifying the agent of the master node's inability to retransmit the messages. The agent then resets the flags corresponding to those messages to indicate that retransmission request(s) need to be sent again.

Abstract: A distributed key-value storage system may include a master node. The key-value store may be distributed among first and second nodes. The master node may receive a publish request to publish one or more key-value pairs. Each key-value pair may be stored in a retransmit buffer and sent to all the first nodes using a communication protocol of a first kind that does not include a retransmit protocol mechanism. Some of the key-value pairs may be sent to one or more second node using a communication protocol of a second kind that includes a retransmit protocol mechanism.

Abstract: An example method is provided to perform policy-based data placement in a virtualized computing environment. The method may comprise receiving a request to create a virtual disk for a virtual machine in the virtualized computing environment. The request may include a usage type associated with the virtual disk. The method may further comprise, based on a data placement policy associated with the usage type, determining a datastore on which to place the virtual disk and to store data of the usage type associated with the virtual disk.

Abstract: Techniques are disclosed for message retransmission. In one embodiment, an agent in a distributed system that receives and/or relays messages from a master node identifies messages that have not been successfully delivered, and adds a stub for each such message to a message queue. The agent then requests retransmission of missing messages in the message queue and sets flags associated with the message stubs to indicate that retransmission requests have been sent. If one (or more) of the messages cannot be retransmitted, the master node sends acknowledgment message(s) notifying the agent of the master node's inability to retransmit the messages. The agent then resets the flags corresponding to those messages to indicate that retransmission request(s) need to be sent again.

Abstract: A method of retransmitting a multicast message through a unicast channel in a virtual distributed storage system comprising a cluster of nodes that includes a master node and a set of agent nodes is provided. Each node maintains a copy of a directory of the virtual distributed storage system. The master node multicasts messages to the agent node through a multicast channel to update the directory. The method determines that a particular message to update the directory exceeds a maximum size limit for transmit to the agent nodes through the multicast channel. The method multicasts a retransmission trigger message from the master node to each agent node. The method receives a retransmission request message at the master node from each of the agent nodes through a unicast channel. The method sends the particular message to update the directory from the master node to each agent node through the unicast channel.

Abstract: An example method is provided to perform policy-based data placement in a virtualized computing environment. The method may comprise receiving a request to create a virtual disk for a virtual machine in the virtualized computing environment. The request may include a usage type associated with the virtual disk. The method may further comprise, based on a data placement policy associated with the usage type, determining a datastore on which to place the virtual disk and to store data of the usage type associated with the virtual disk.