Aneka Cloud

ApplicationPlatform
Introduction
• Aneka consists of a scalable cloud middleware that can be deployed
on top of heterogeneous computing resources.
• It offers an extensible collection of services coordinating the
execution of applications, helping administrators monitor the status
of the cloud, and providing integration with existing cloud
technologies.
• One of Aneka’s key advantages is its extensible set of application pro-
gramming interfaces (APIs) associated with different types of
programming models—such as Task, Thread, and MapReduce—used
for developing distributed applications, integrating new capabilities
into the cloud, and supporting different types of cloud deployment
models: public, private, and hybrid
• Aneka is a pure PaaS solution for cloudcomputing
Framework:
Services: Fabric Services, Foundation Services, and Execution Services.
These take care of infrastructure management, supporting services
for
Aneka Cloud and Application Management and Execution respectively.
These services are made available by means of the application
management and development layer, which includes interfaces and
APIs for developing cloud applications and the management tools and
interfaces for controlling Aneka Clouds.
The framework includes the basic services for infrastructure and node
management, application execution, accounting, and system
monitoring
• Elasticity and scaling - By means of the dynamic provisioning service,
Aneka supports dynamically upsizing and downsizing of the
infrastructure available for applications.
• Run time management. The runtime machinery is responsible for
keeping the infrastructure up and running and serves as a hosting
environment for services. It is primarily represented by the container
and a collection of services that manage service membership and
lookup, infrastructure maintenance, and profiling.
• Resource management. - Aneka is an elastic infrastructure in which
resources are added and removed dynamically according to
application needs and user requirements. To provide QoS-based
execution, the system not only allows dynamic provisioning but also
provides capabilities for reserving nodes for exclusive use by specific
applications
Aneka container can be classified into three
major categories:
• Fabric Services
• Foundation Services
• Application Services
The services stack resides on top of the Platform Abstraction Layer(PAL),
representing the interface to the underlying operating system and
hardware. PAL is responsible for:
• Uniform and platform-independent implementation interface for accessing
the hosting platform
• Uniform access to extended and additional properties of the hosting
platform
• Uniform and platform-independent access to remote nodes
• Uniform and platform-independent management interfaces
More information:
The collectible data that are exposed by the PAL are the following:
• Number of cores, frequency, and CPU usage
Memory size and usage
• Aggregate available disk space
• Network addresses and devices attached to the node
Fabric Services
• Fabric Services define the lowest level of the software stack
representing the Aneka Container. They provide access to the
resource-provisioning subsystem and to the monitoring facilities
implemented in Aneka.
• Resource-provisioning services are incharge of dynamically providing
new nodes on demand by relying on virtualization technologies,while
monitoring services allow for hardware profiling and implement a
basic monitoring infrastructure that can be used by all the services
installed in the container.
Profiling and monitoring
• Profiling and monitoring services are mostly exposed through the
Heartbeat Monitoring, and Reporting Services. The first makes
available the information that is collected through the PAL;

• the other implement a generic infrastructure for monitoring the

activity of any service in the Aneka Cloud.
Resource Management
• It comprises several tasks: resource membership, resource
reservation , and resource provisioning.
• Aneka provides a collection of services that are incharge of managing
resources. These are the Index Service (or Membership Catalogue),
Reservation Service, and Resource Provisioning Service.
• The Membership Catalogue is Aneka’s fundamental component for
resource management; it keeps track of the basic node information
for all the nodes that are connected or disconnected.
• The Membership Catalogue implements the basic Services of a
directory service, allowing the search for services using attribute such
as names and nodes
• The Membership Catalogue is also the collector of the dynamic
performance data of each node, which are then sent to the local
monitoring service to be persisted in the long term.
• Dynamic resource provisioning allows the integration and
management of virtual resources leased from IaaS providers into the
Aneka Cloud. This service changes the structure of the Aneka Cloud
by allowing it to scale up and down according to different needs:
handling node failures, ensuring the quality of service for
applications, or maintaining a constant performance and throughput
of the Cloud.
Foundation services
• Foundation Services are related to the logical management of the
distributed system built on top of the infrastructure and provide
supporting services for the execution of distributed applications.
• These services cover: • Storage management for applications -a
centralized file storage, which is mostly used for the execution of
compute- intensive applications, and a distributed file system, which
is more suitable for the execution of data-intensive applications.
• Concept : file channel that identifies a pair of components: a file
channel controller and a file channel handler.
• • Accounting, billing, and resource pricing
• • Resource reservation
• Centralized storage is implemented through and managed by Aneka’s
Storage Service. The service constitutes Aneka’s data-staging
facilities. It provides distributed applications with the basic file
transfer facility.
• To support different protocols, the system introduces the concept of
a file channel that identifies a pair of components :a file channel
controller and a file channel handler.
• The file channel controller constitutes the server component of the
channel, where files are stored and made available
• The file channel handler represents the client component, which is
used by user applications or other components of the system to
upload,download,or browse files.
Distributed File System
• The architecture of the file system is based on a master node, which
contains a global map of the file system and keeps track of the status
of all the storage nodes, and a pool of chunk servers, which provide
distributed storage space in which to store files.
• Files are huge by traditional standards (multi-gigabytes).
• Files are modified by appending new data rather than rewriting
existing data.
• There are two kinds of major workloads: large streaming reads and
small random reads.
• It is more important to have a sustained bandwidth than a low
latency.
Accounting, billing, and resource pricing
• Accounting Services keep track of the status of applications in the
Aneka Cloud.
• The collected information provides a detailed breakdown of the
distributed infrastructure usage and is vital for the proper
management of resources.
• Aneka Billing Service provides detailed information about each user’s
usage of resources, with the associated costs. Each resource can be
priced differently according to the set of services that are available on
the corresponding Aneka container
Resource reservation
• Aneka’s Resource Reservation supports the execution of distributed
applications and allows for reserving resources for exclusive use by
specific applications.
• Two steps: Resource Reservation and the Allocation Service.
• The Allocation Service is installed on each node that features
execution services and manages the database of information
regarding the allocated slots on the local node.
• If it is possible to satisfy the request, the Reservation Service will
return a reservation identifier as proof of the resource booking.
During application execution, such an identifier is used to select the
nodes that have been reserved, and they will be used to execute the
application.
framework supports three different
implementations:
• Basic Reservation- Features the basic capability to reserve execution
slots on nodes and implements the alternate offers protocol, which
provides alternative options incase the initial reservation request
cannot be satisfied.
• Libra Reservation. features the ability to price nodes differently
according to their hardware capabilities.
• Relay Reservation. Constitutes a very thin implementation that allows
are source broker to reserve nodes in Aneka Clouds and control the
logic with which these nodes are reserved. This implementation is
useful in integration scenarios in which Aneka operates in an intercloud
environment.
Application Services
• Application Services :manage the execution of applications and
constitute a layer that differentiates according to the specific
programming model used for developing distributed applications on
top of Aneka.
• Aneka defines a reference model for implementing the runtime
support for programming models that abstracts these two activities
in corresponding services: the Scheduling Service and the Execution
Service.
• Aneka does not provide a centralized scheduling engine, but each
programming model features its own scheduling service that needs to
work in synergy with the existing services of the middle- ware.
• In this scenario, common situations that have to be appropriately
managed are the following: multiple jobs sent to the same node at
the same time; jobs without reservations sent to reserved nodes; and
jobs sent to nodes where the required services are not installed.
• The current design philosophy in Aneka is to keep the scheduling
engines completely separate from each other and to leverage existing
services when needed. As a result, it is possible to enforce that only
one job per programming model is run on each node at any given
time
Private Cloud Infrastructure
Public Cloud Deployment
Aneka SDK
Application Model
• Each distributed application running on top of Aneka is an instance of
the ApplicationBase <M> class, where M identifies the specific type of
application manager used to control the application.
• Application classes provides a view to the developers
• Applications – set of tasks
• Aneka further specializes applications into two main categories:
(1) applications whose tasks are generated by the user and
(2) applications whose tasks are generated by the runtime
infrastructure.
Service Model
Management Tools
• Infrastructure Management
• Platform Management
• Application Mangement