Facebook, Inc.

Future Wei Technologies, Inc.

Huawei Technologies Co., Limited

Jianming Wu

Guangyu Shi

Arun Dattaram Sharma

Chunhui Zhu

Masood Mortazavi

Stephen Charles Heise

Byung Choi

Alessandro Presta

Fu Li

Avraham Kapuya

Technology is disclosed for establishing a querying system and load balancing the multi-tiered querying system. A multi-tiered targeted query system can comprise three tiers: a web tier, an aggregator tier, and a shard tier. When load balancing of shards is performed, fan-out can occur, increasing latency. The disclosed technology performs load balancing while minimizing the amount fan-out increase. Selecting the databases to move can comprise determining which databases on that server are causing the most load, and determining if moving any of these databases will increase an expected amount of fan-out above an acceptable threshold value. Determining the expected amount of fan-out increase incurred by moving a database can be based on an analysis of a number of friend relationships between that database and other databases on the same or other servers.

Load balancing for multi-tiered querying

Technology is disclosed for a multi-tiered querying system to target queries to systems storing data relevant to the query. A multi-tiered targeted query system comprises at least three tiers: a web tier, an aggregator tier, and a shards tier. Servers at the web tier can be configured to service user data requests and pass them to servers at the aggregator tier. Servers at the aggregator tier can be configured to determine which selected shard servers have the requested information; formulate queries for the selected shard servers; send the queries to the selected shard servers; and aggregate results from the selected shard servers. Servers at the shard tier can be configured to store data, receive queries on that data, and return results for received queries.

Multi-tiered targeted querying

Unique identifiers (IDs) associated with a plurality of nodes may be provided. Nodes clustered within a community may be assigned numerically proximate unique IDs. A number of partitions associated with a plurality of machines may be determined. The unique IDs may be segmented into divisions based on the number of partitions. The unique IDs may be mapped to the plurality of machines based on the divisions.

Systems and methods for mapping and routing based on clustering

A method performed by a switch, the method comprising receiving a request for a file from a client, wherein the file is located in a distributed storage system, extracting a directory for the file from the request, performing longest prefix matching (LPM) of the directory against a forwarding information base (FIB) to identify a port, wherein the FIB associates directories with ports of the switch, and forwarding the client request toward a server containing the file via the identified port, wherein the server is one of a plurality of servers in the distributed storage system.

Scalable storage systems with longest prefix matching switches

A method for determining a location of a requested resource, comprising receiving a request for a resource, wherein the resource is identified by a hierarchically structured name, decomposing the hierarchically structured name into a plurality of segments, comparing each of the plurality of segments to parent nodes of a plurality of two-level treelets, wherein the comparing of at least two of the plurality of segments to parent nodes of the plurality of two-level treelets is performed in a parallel manner, wherein combinations of two-level treelets comprise a hierarchical name of a network resource, and determining a longest prefix match of the hierarchically structured name.

Method and system for longest prefix matching of variable-sized hierarchical names by treelets

A router forwarding plane comprising a bloom filter stored on a first tier storage medium, and a forwarding information log associated with the bloom filter and stored on a second tier storage medium. Also disclosed is a network component comprising a receiver configured to receive a content comprising a general name prefix, a first tier storage medium configured to store a plurality of bloom filters associated with a plurality of general name prefixes and a plurality of corresponding ports, a logic circuitry configured to compute a plurality of signatures based on the general name prefix of the received content, and a transmitter configured to forward the received content on at least one of the ports that are associated with at least one of the bloom filters if the general name prefix is a member of the at least one of the bloom filters.

Content router forwarding plane architecture

A key/value store comprising a first tier storage device configured to store information about a plurality of keys for a plurality of values without the values, and a second tier storage device coupled to the first tier storage device and configured to store the values associated with the keys without the keys, wherein the first tier storage device has lower latency and higher throughput than the second tier storage device, and wherein the second tier storage device has higher capacity than the first tier storage device. Also disclosed is a method comprising receiving a key/value operation request at a first tier storage device, mapping a key in the key/value operation request to a locator stored in a second tier storage device if the key/value operation request is valid, and mapping the locator to a value in a third tier storage device if the key has a corresponding locator.

Growing community of inventors

Fremont, CA, United States of America

Jianming Wu