The Java Multi-Cloud Toolkit
Go to file
adriancole 992e9ebbcf change cacheloader that can return nulls to return optional so to avoid cacheload exceptions 2013-01-10 13:20:27 -08:00
all Moved openstack-cinder and rackspace-cloudblockstorage-us/uk out of labs. 2012-11-13 18:48:30 -06:00
allblobstore issue #1107: remove walrus and related providers 2012-12-29 12:00:49 -08:00
allcompute Merge pull request #1116 from jclouds/remove-rimuhosting 2012-12-29 17:42:00 -08:00
allloadbalancer Renamed cloudloadbalancers to rackspace-cloudloadbalancers. 2012-12-06 14:42:48 -06:00
apis change cacheloader that can return nulls to return optional so to avoid cacheload exceptions 2013-01-10 13:20:27 -08:00
archetypes replaced @Timeout w/explicit properties 2012-12-30 12:31:43 -08:00
assemblies Remove Slicehost litter 2013-01-08 12:05:40 -08:00
blobstore removed custom guava code for scoped binding and type params 2013-01-08 22:48:10 -08:00
common removed custom guava code for scoped binding and type params 2013-01-08 22:48:10 -08:00
compute ensure extra metadata from provider doesn't trip metadata tests 2013-01-01 17:18:55 -08:00
core refactored tests and internal code due to removing custom guava code 2013-01-08 22:23:47 -08:00
demos undid Serializable and added checkstyle InterfaceIsType:allowMarkerInterfaces=false 2012-11-18 09:23:44 -08:00
drivers First wave of RestAnnotationParser refactoring introduces Invokable, removes SeedAnnotationCache and associated race conditions 2013-01-04 13:51:46 -08:00
labs removed custom guava code for scoped binding and type params 2013-01-08 22:48:10 -08:00
loadbalancer undid Serializable and added checkstyle InterfaceIsType:allowMarkerInterfaces=false 2012-11-18 09:23:44 -08:00
project Remove explicit dependency on bouncycastle for PEM parsing 2012-12-28 23:11:42 -08:00
providers change cacheloader that can return nulls to return optional so to avoid cacheload exceptions 2013-01-10 13:20:27 -08:00
resources undid Serializable and added checkstyle InterfaceIsType:allowMarkerInterfaces=false 2012-11-18 09:23:44 -08:00
sandbox-apis dependent changes on tests and internal classes 2013-01-08 00:32:45 -08:00
sandbox-drivers/asynchttpclient Enforce newline at end of file via Checkstyle 2012-10-29 22:12:44 -07:00
sandbox-providers dependent changes on tests and internal classes 2013-01-08 00:32:45 -08:00
scriptbuilder added literal statement for when you don't want any magic 2013-01-01 17:08:17 -08:00
skeletons ensure extra metadata from provider doesn't trip metadata tests 2013-01-01 17:18:55 -08:00
.gitignore added support for vbox 4.2.6; 2012-12-31 01:33:43 +01:00
.mailmap .mailmap updates 2012-10-31 15:08:04 -07:00
README.md Use Java and Clojure syntax highlighting in the readme file. 2012-12-09 22:04:23 +03:00
pom.xml Issue 1112:Move antcontrib to new jclouds-ant repository 2012-10-21 15:07:18 -07:00

README.md

jclouds

jclouds allows provisioning and control of cloud resources, including blobstore and compute, from Java and Clojure. Our API gives allows developers to use both portable abstractions and cloud-specific features. We test support of dozens of cloud providers and cloud software stacks, including Amazon, Azure, GoGrid, Ninefold, OpenStack, and vCloud. jclouds is licensed under the Apache License, Version 2.0

Features

Even if you don't need the portable apis we provide, or could roll it your own, programming against cloud environments can be challenging. We focus on the following areas so that you can focus on using the cloud, rather than troubleshooting it!

  • SIMPLE INTERFACE Instead of creating new object types, we reuse concepts like maps so that the programming model is familiar. In this way, you can get started without dealing with REST-like apis or WS.

  • RUNTIME PORTABILITY We have drivers that allow you to operate in restricted environments like Google App Engine. We have very few required dependencies, so we are unlikely to clash with your app.

  • DEAL WITH WEB COMPLEXITY Network based computing introduces issues such as transient failures and redirects. We handle this for you.

  • UNIT TESTABILITY Writing tests for cloud endpoints is difficult. We provide you with Stub connections that simulate a cloud without creating network connections. In this way, you can write your unit tests without mocking complexity or the brittleness of remote connections.

  • PERFORMANCE Writing tests for cloud endpoints is difficult. We provide you with Stub connections that simulate a cloud without creating network connections. In this way, you can write your unit tests without mocking complexity or the brittleness of remote connections.

  • LOCATION All of our abstractions are location-aware. For example, you can get ISO-3166 codes to tell which country or province a cloud runs in.

  • QUALITY We test every provider with live scenarios before each release. If it doesn't pass, the provider goes into the sandbox.

BlobStore

Simplifies dealing with key-value providers such as Amazon S3. For example, BlobStore can give you a simple Map view of a container.

BlobStore Example (Java):

// init
context = new BlobStoreContextFactory().createContext(
"aws-s3",
accesskeyid,
secretaccesskey);
blobStore = context.getBlobStore();

// create container
blobStore.createContainerInLocation(null, "mycontainer");

// add blob
blob = blobStore.blobBuilder("test").payload("testdata").build();
blobStore.putBlob("mycontainer", blob);

BlobStore Example (Clojure):

(use 'org.jclouds.blobstore2)
(def *blobstore* (blobstore "azureblob" account encodedkey))
(create-container *blobstore* "mycontainer")
(put-blob *blobstore* "mycontainer" (blob "test" :payload "testdata"))

ComputeService

Simplifies the task of managing machines in the cloud. For example, you can use ComputeService to start 5 machines and install your software on them.

Compute Example (Java):

// init
context = new ComputeServiceContextFactory().createContext(
"aws-ec2",
accesskeyid,
secretaccesskey,
ImmutableSet.of(new Log4JLoggingModule(), new SshjSshClientModule()));

client = context.getComputeService();

// define the requirements of your node
template = client.templateBuilder().osFamily(UBUNTU).smallest().build();

// setup a boot user which is the same as your login
template.getOptions().runScript(AdminAccess.standard());

// these nodes will be accessible via ssh when the call returns
nodes = client.createNodesInGroup("mycluster", 2, template);

// you can now run ad-hoc commands on the nodes based on predicates
responses = client.runScriptOnNodesMatching(inGroup("mycluster"), "uptime", wrapInInitScript(false));

Compute Example (Clojure):

(use 'org.jclouds.compute2)

; create a compute service using sshj and log4j extensions
(def compute (*compute* "trmk`-ecloud" "user" "password" :sshj :log4j))

; launch a couple nodes with the default operating system, installing your user.
(create-nodes *compute* "mycluster" 2
(TemplateOptions$Builder/runScript (AdminAccess/standard)))

; run a command on that group 
(run-script-on-nodes-matching *compute* (in-group? "mycluster") "uptime" 
(RunScriptOptions$Builder/wrapInInitScript false))

Check out https://github.com/jclouds/jclouds-examples for more examples!

Downloads

Resources

License

Copyright (C) 2009-2012 jclouds, Inc.

Licensed under the Apache License, Version 2.0