Bhojpur Cache - Management Engine
The Bhojpur Cache is a high-performance data caching platform applied within Bhojpur.NET Platform for delivering web-scalable applications or services. It could utilize different kinds of data storage engines (e.g., in-memory, file-based) depending on the application's use cases.
Key Features
- Multi-modal Storage Engines
- ACID transactions
- Web Services APIs
Getting Started
To install Bhojpur Cache, use the go get
command:
$ go get github.com/bhojpur/cache/...
Introspection Dashboard
To debug the Bhojpur Cache, you can add an
introspection handler to an HTTP mux and get bettervisibility into in-memory
database storage engine's behaviour. For example
$ go build -o bin/cachedbg ./internal/main.go
http.Handle("/introspect", http.StripPrefix("/introspect", debugger.NewHandler(mydb)))
then, run the bin/cachedbg
binary by passing in the path to your database:
$ bin/cachedbg ./internal/path/to/my.db
After pointing your web browser to http://localhost:3000
, you should see something like this

It allows you to introspect Bhojpur Cache
database in a web browser. The bin/cachedbg
tool gives you access to
low-level page information and b-tree structures so you can better understand
how Bhojpur Cache is laying out your data.
HTTP Integration
You can also use boltd as an http.Handler
in your own application. To use it,
simply add the handler to your muxer:
To generate a custom web template, you need the following tool
$ go get github.com/benbjohnson/ego
Distributed Applications