This interface defines pure-virtual methods for each of the user-facing overload sets in Table. This allows users to inject custom behavior (e.g., with a Google Mock object) in a Table object for use in their own tests.
This class opens a number of gRPC Channels upon its construction. These channels each initiate a connection to the Cloud Bigtable service. This is a relatively slow operation that can take milliseconds, so applications are advised to reuse DataConnection objects when possible.
The exact number of channels can be configured with the google::cloud::GrpcNumChannelsOption. If this option is not set, the class creates between 1 and 64 channels. The specific algorithm and number of channels is an implementation detail, and subject to change without notice.
Each request sent from the client library cycles through the channels in the connection pool in a round robin fashion.
Unused gRPC channels can enter an idle state. Refreshing them during a request may incur additional latency. To avoid this, the client library keeps connections open by refreshing the channels in a background thread. Advanced users can configure the frequency of the refreshes, or disable this feature, by using bigtable::MinConnectionRefreshOption and bigtable::MaxConnectionRefreshOption.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2024-12-16 UTC."],[],[]]
