Abstract: Disclosed is an approach for implementing a metadata cache in a virtualization system. A self-adaptive approach is provided to keep compressed and uncompressed entries together in cache. Along with adaptive nature, disclosed is an approach to prioritize critical workloads for the cache.

Abstract: Various embodiments set forth techniques for cache warmup. The techniques determining, by a node, identities of one or more target storage blocks of a plurality of storage blocks managed by a storage system, where the node previously cached metadata corresponding to the one or more target storage blocks; receiving the metadata corresponding to the one or more target storage blocks; and storing the metadata corresponding to the one or more target storage blocks in a cache memory of the node.

Abstract: Disclosed is an approach for implementing a metadata cache in a virtualization system. A self-adaptive approach is provided to keep compressed and uncompressed entries together in cache. Along with adaptive nature, disclosed is an approach to prioritize critical workloads for the cache.

Abstract: In various embodiments, a computer-implemented method for supporting consistent secondary indexes, comprises receiving, at a first node, a write request comprising a data entry, storing the data entry in an in-memory structure separate from a primary structure for storing the data entry, generating, based on the data entry, a secondary index data entry for a secondary index, and transmitting the secondary index data entry to a second node for inclusion in the secondary index.

Abstract: Various embodiments set forth techniques for cache warmup. The techniques determining, by a node, identities of one or more target storage blocks of a plurality of storage blocks managed by a storage system, where the node previously cached metadata corresponding to the one or more target storage blocks; receiving the metadata corresponding to the one or more target storage blocks; and storing the metadata corresponding to the one or more target storage blocks in a cache memory of the node.

Abstract: Methods, systems, and computer program products for performing inline garbage collection of the portion of a data storage area corresponding to a to-be-deleted key of a key-value pair.

Abstract: An apparatus includes a processor having programmed instructions to determine a container number of container instances to be deployed in a cluster based on compute resources and determine a node number of virtual nodes to be deployed in the cluster based on storage resources. The node number of virtual nodes includes a key-value store. Each of the node number of virtual nodes owns a corresponding key range of the key-value store. The processor has programmed instructions to distribute the node number of virtual nodes equally across the container number of container instances and deploy the container number of container instances.

Abstract: An apparatus includes a processor having programmed instructions to determine a container number of container instances to be deployed in a cluster based on compute resources and determine a node number of virtual nodes to be deployed in the cluster based on storage resources. The node number of virtual nodes includes a key-value store. Each of the node number of virtual nodes owns a corresponding key range of the key-value store. The processor has programmed instructions to distribute the node number of virtual nodes equally across the container number of container instances and deploy the container number of container instances.

Abstract: Systems for distributed computing systems. A topological configuration of computing nodes is selected to manage availability of metadata data in a computing system. A method embodiment commences by accessing a plurality of node topology attributes and using those attributes to map between nodes and availability domains. Resource usage measurements such as computing node load are collected. A plurality of candidate replication configurations are generated, and each candidate replication configuration is scored with respect to several quantitative objectives. Additionally, the candidate replication configurations are given respective resource usage balance scores. One or more candidate replication configurations are selected based on resource usage balance scores and/or a separation skew value. Determination of a selected configuration is dominated by resource usage when there is a tie between best-scoring configurations or when none of the configurations meet a scoring threshold.

Abstract: The system can include a cache and cluster manager. The cache can store a plurality clusters, each of a plurality of clusters including a plurality of cache entries, each of the plurality of cache entries including a plurality of first metadata feature values. The cluster manager can assign a first cache entry corresponding to a data record located in memory to a first cluster based on determining a lowest distance. The lowest distance is determined by operations. The operations can include calculating a plurality of intra cluster feature means. The operations can include receiving a plurality of second metadata feature values of the first cache entry. The operations can include calculating a plurality of distances based on the plurality of intra cluster feature means and the plurality of second metadata feature values. The operations can include determining the first entry having a lowest distance of the plurality of distances.

Abstract: The system can include a cache and cluster manager. The cache can store a plurality clusters, each of a plurality of clusters including a plurality of cache entries, each of the plurality of cache entries including a plurality of first metadata feature values. The cluster manager can assign a first cache entry corresponding to a data record located in memory to a first cluster based on determining a lowest distance. The lowest distance is determined by operations. The operations can include calculating a plurality of intra cluster feature means. The operations can include receiving a plurality of second metadata feature values of the first cache entry. The operations can include calculating a plurality of distances based on the plurality of intra cluster feature means and the plurality of second metadata feature values. The operations can include determining the first entry having a lowest distance of the plurality of distances.

Abstract: Systems for distributed computing systems. A topological configuration of computing nodes is selected to manage availability of metadata data in a computing system. A method embodiment commences by accessing a plurality of node topology attributes and using those attributes to map between nodes and availability domains. Resource usage measurements such as computing node load are collected. A plurality of candidate replication configurations are generated, and each candidate replication configuration is scored with respect to several quantitative objectives. Additionally, the candidate replication configurations are given respective resource usage balance scores. One or more candidate replication configurations are selected based on resource usage balance scores and/or a separation skew value. Determination of a selected configuration is dominated by resource usage when there is a tie between best-scoring configurations or when none of the configurations meet a scoring threshold.

Abstract: An exemplary embodiment herein is a method including comparing a cache hit rate ratio of a unified cache to a first pre-determined threshold, incrementing an alert counter in response to the cache hit rate ratio being lower than the first pre-determined threshold, comparing the alert counter to a pre-determined limit, preventing a first receipt of a type of data by the unified cache in response to the alert counter being equal to the pre-determined limit, causing a second receipt of metadata by the unified cache, comparing the cache hit rate ratio to a second pre-determined threshold, and allowing the first receipt of the type of data by the unified cache in response to the cache hit rate ratio being greater than the second pre-determined threshold.

Abstract: An exemplary embodiment herein is a method including comparing a cache hit rate ratio of a unified cache to a first pre-determined threshold, incrementing an alert counter in response to the cache hit rate ratio being lower than the first pre-determined threshold, comparing the alert counter to a pre-determined limit, preventing a first receipt of a type of data by the unified cache in response to the alert counter being equal to the pre-determined limit, causing a second receipt of metadata by the unified cache, comparing the cache hit rate ratio to a second pre-determined threshold, and allowing the first receipt of the type of data by the unified cache in response to the cache hit rate ratio being greater than the second pre-determined threshold.