Piranha is the IP Load Balancing solution in Redhat Enterprise Linux. Piranha provides the ability to load-balance incoming IP network requests across a pool of servers. IP Load Balancing with Piranha is based on open source Linux Virtual Server (LVS) technology, with significant Red Hat enhancements. Piranha is strictly a software HA implementation which is the cluster package and the GUI administrative tool for the cluster.

The Piranha IP Load Balancing cluster consists of at least two layers. The first layer is composed of a pair of similarly configured Red Hat Enterprise Linux AS or ES systems with Red Hat Cluster Suite installed. One of these nodes acts as the active IPLB router while the other acts as a backup or standby. The active Load Balancer directs requests from the clients to the second layer, a pool of servers called real servers. The real servers provide the critical services to the end-user while the LVS router balances the load to these servers.

Requests for service from a Piranha cluster are sent to a virtual server address: a unique triplicate consisting of a protocol, a floating IP address, and a port number. A router (IP Load Balancer) receives job requests and redirects those requests the real servers that actually process the requests.

The active router serves two roles in the cluster:

• Balancing the load on the real servers.
• Checking integrity of the services on each of the real servers.

The simple, two-layered configuration is best for clusters serving static data that does not change very frequently like static Web pages. If high-availability shared data access and update is required, a third layer is added to the topology.

The Piranha clustering system includes the following components:

IP Virtual Server (IPVS)

lvs daemon to manage the IPVS routing table via the ipvsadm tool (LVS).

nanny daemon to monitor servers and services on real servers in a cluster.

pulse daemon to control the other daemons and handle failovers between IPVS routing boxes.

piranha GUI tool to administer and manage the cluster environment.

All of these daemons use the same configuration file, named /etc/lvs.cf by default. Piranha's primary function is to start and stop pulse interactively and to edit the contents of the /etc/lvs.cf config file.

The IPVS code provides the controlling intelligence. It matches incoming network traffic for defined virtual servers and redirects each request to a real server, based on an adaptive scheduling algorithm. The scheduler supports two classes of scheduling, each with a weighted and non-weighted version. There is a basic Round Robin scheduler that simply rotates between all active real servers. The more complex scheduler, Least Connections, keeps a history of open connections to all real servers and sends new requests to the real server with the least number of open connections.

The weighted versions of those two schedulers allow the administrator to mix heterogeneous hardware and OSes as the real servers and have the load redistribution reflect physical differences between the machines. As an additional benefit, the Piranha clustering package will adaptively modify this weight based on the load averages of the real servers.

IPVS supports three types of network configurations. Network Address Translation (NAT), Tunneling and Direct Routing. NAT requires that there be a public address for the virtual server(s) and a private subnet for the real servers. It then uses IP Masquerading for the real servers. Tunneling uses IP encapsulation and reroutes packets to the real servers. This method requires that the real servers support a tunneled device to unencode the packets. Direct routing rewrites the IP header information and then resends the packet directly to the real server.

Each service running on a real server being routed to as a part of a virtual server is monitored by a nanny process running on the active IPVS router. These service monitors follow a two-step process. First, the hardware/network connectivity is checked to ensure that the real server is responding to the network. Second, a connect is sent to the port of the real server that has the monitored service running on it. Once connected, nanny sends a short header request string and checks to make sure that it receives a banner string back. This process is repeated every two seconds. If a sufficient time (configurable) elapses with no successful connects, the real server is assumed dead and is removed from the IPVS routing table. Nanny continues to monitor the real server and when the service has returned and has remained alive for a specified amount of time, the server's place in the IPVS routing table is restored.

The IPVS router is a single point of failure (SPOF) so support for a hot standby node is supported. When configured with a standby, the inactive machine maintains a current copy of the cluster's configuration file (/etc/lvs.cf) and heartbeats across the public network between it and the active IPVS router node. If, after a specified amount of time, the active router fails to respond to heartbeats, the inactive node will execute a failover. The failover process consists of recreating the last known IPVS routing table and stealing the virtual IP(s) that the cluster is responsible for. It also steals the private gateway IP for the real servers. It brings up those IPs on its public and private networks, then sends out gratuitous ARPs to announce the new MAC addresses for the relocated IPs. It also starts up the appropriate nanny process to monitor services on the real servers. Should the failed node return to life, it will announce its return in the form of heartbeats and will become the new inactive hot standby IPVS router.

Ultra Monkey is a project to create load balanced and highly available network services. UltraMonkey is an opensource project that uses Linux Virtual Server and Heartbeat opensource projects.

For example a cluster of web servers that appear as a single web server to end-users. The service may be for end-users across the world connected via the internet, or for enterprise users connected via an intranet.

An end-user sends a packet to the virtual service, Ultra Monkey selects a server and forwards the packet. When the server replies, Ultra Monkey forwards this reply back to the end-user. This is the very basic operation of Ultramonkey.

Ultra Monkey makes runs on Linux operating system to provide a flexible solution that can be tailored to a wide range of needs. From small clusters of only two nodes to large systems serving thousands of connections per second.

While the Ultra Monkey software itself runs on Linux, it is able to provide clustering for virtually any network services running on an operating system that can communicate using TCP or UDP. This means almost any operating system, including Windows™, Solaris™ and of course Linux itself and a wide range of protocols, with native health checks for: Web, Mail, FTP, News, LDAP and DNS. Other protocols can be suported using a non-native health check or by adding a custom health check.

UltraMonkey-L7

UltraMonkey-L7 project extends from the earlier Ultra Monkey project and provides load balancing features at Layer 7. The architecture is flexible to incorporate various protocols and load balancing methods, that can be added in the future.

Over the recent years, SSL encryption is used to prevent loss of information during communication in web applications. The architecture of UltraMonkey-L7 has been enabled to easily incorporate the processing of SSL communication.

Key Features

• Layer 7 switching using the Linux Virtual Server.
• Load-balancing using specific character string interpretation of URL
• Session persistence using specific character string interpretation of URLEasily expandable to a large number of IP based virtual services.
• High Availability provided by Heartbeat protocol.
• Supports Highly Available and Load Balanced topologies with working configuration examples. 

For more information and download, click here

Eddie is a high availability clustering tool. It is an open source, 100% software solution written primarily in the functional programming language Erlang (http://www.erlang.org/) and is available for Solaris, Linux and *BSD.

Eddie provides advanced automatic traffic management and configuration of geographically distributed server sites, consisting of one or more Local Area Networks.

At each site, certain servers are designated as Front End Servers. These servers are responsible for controlling and distributing incoming traffic across designated Back End Servers, and tracking the availability of Back End Web Servers within the site. Back End Servers may support a range of Web servers, including Apache.

Currently, Eddie consists of two main software packages:

Enhanced DNS

For server sites distributed over a city, a country, or even the world, Eddie's Enhanced DNS allows the entire distributed web server capacity to be usable. The individual sites may have different numbers of servers, different vintages of computer, even different operating systems.

The load information passed between servers at a site is processed locally and summary information passed to each Eddie Enhanced DNS server, across the Internet if necessary.

The Eddie Enhanced Domain Name Service package dynamically balances client domain name resolution requests across all sites in the distributed server, and more specifically, across each Front End within each site. The balancing is performed to keep each site working at approximately the same fraction of its overall capacity, rather than equalising the number of requests sent to each site, as occurs with Round Robin based schemes. This ensures maximum efficiency of Front End Server resources within each site.

Intelligent HTTP Gateway

Within a single site, Eddie's Intelligent HTTP Gateway supports superior functionality and performance.  Load information passed to Front End Servers running the Intelligent HTTP Gateway package enables,

• Dynamic load balancing to be employed, maximising the capacity of the site.
• Automatic discovery of failed servers and subsequent re-direction of traffic, thereby improving reliability.
• Site scalability through simple addition of extra servers as needed.
• Advanced admission controls to be employed to improve user Quality of Service.

For more information and download, please click here

Balance from inLab Software GmBH is a simple, generic TCP proxy, which allows simple round-robin load balancing and graceful failover between several destination servers.

Balance is available at http://balance.sourceforge.net/.

A possible destination consisting of a host address and a port is called a "channel". A channel is member of a "channel group". Channels are numbered in a group starting with 0. Groups are numbered starting with 0, which is the initial default group. Balance accepts connections on the given port and forwards them to the supplied channels. At least one channel (in the default group) must be specified. If there are two or more channels specified in a group balance performs a simple round-robin load balancing between the channels.

Balance allows the definition of further channel groups. The connection scheme works as follows:

balance tries first to establish a connection to a channel in the first group (0), performing the standard round-robin load balancing scheme. If no channel in this group is available, balance proceeds with the next higher channel group. Groups are simply separated with a "!" at the command line at startup and can be controlled interactively with the "group" command.

A "%" instead of a "!" as a group separator declares the previous group to be of type "hash". This means that instead of a round-robin algorithm, a hash distribution based on the client ip address is used to determine the destination channel. This allows connecting one client always to the same server (e.g. balancing http sessions to a single server). Hosts may be specified either by hostname or by IP address. Ports may be specified either by name (as listed in /etc/services) or numerically. If no port is specified in a destination, the destination port defaults to the source port that balance controls.

Balance allows the specification of the maximum number of connections per channel. This parameter can be optionally added after the port specification separated by a colon (":"). If a maximum number of connections is specified a channel will only be used for this maximum number of simultaneous connections.

A maxc value of 0 denotes an unlimited number of connections. This is the initial default value of a channel.

The maximum number of groups and channels balance can handle is specified at compile time and is initially 16 channels in 16 groups.

Failover to another destination (a "channel") occurs if the connection is refused on the current channel or if the connect timeout is reached trying to establish a connection. If all possible destinations (channels) currently fail, the client connection to balance is closed.

For more information, click here

A professional edition of balance, BalanceNG offers much more advanced features including Layer2 (ethernet) load balancing. It is now available under a new Free Basic License: A basic one node load balancer setup with one virtual server and two targets may now be evaluated as long as needed at no charge. 

For more information, click here

The project Linux Virtual Server is aimed to

"Build a high-performance and highly available server for Linux using clustering technology, which provides good scalability, reliability and serviceability."

The LVS cluster system is also known as load balancing server cluster.

The Linux Virtual Server is a highly scalable and highly available server built on a cluster of real servers, with the load balancer running on the Linux operating system. The architecture of the server cluster is fully transparent to end users, and the users interact as if it were a single high-performance virtual server.

The Linux Virtual Server as an advanced load balancing solution can be used to build highly scalable and highly available network services, such as scalable web, cache, mail, ftp, media and VoIP services.

The LVS load balancing here can be a Layer-4 switching (Load Balancing) or a layer-7 switching (Load Balancing).

Layer-4 load balancing is to distribute requests to the servers at transport layer, such as TCP, UDP and SCTP transport protocol. The load balancer distributes network connections from clients who know a single IP address for a service, to a set of servers that actually perform the work. Since connection must be established between client and server in connection-oriented transport before sending the request content, the load balancer usually selects a server without looking at the content of the request. Layer-4 load balancing can also be used to balance traffic at mutliple Internet access links, in order to increase Internet access speed.

IPVS is an implementation of layer-4 load balancing for the Linux kernel and ipvsadm is the utility used to create,list and edit IP Virtual Servers.

Layer-7 load balancing, also known as application-level load balancing, is to parse requests in application layer and distribute requests to servers based on different types of request contents, so that it can provide quality of service requirements for different types of contents and improve overall cluster performance. The overhead of parsing requests in application layer is high, thus its scalability is limited, compared to layer-4 load balancing.

KTCPVS is an implementation of layer-7 load balancing for the Linux kernel. With the appropriate modules, the Apache, Lighttpd and nginx web servers can also provide layer-7 load balancing as a reverse proxy.

tcpvsadm is the utility to admin KTCPVS

TCPSplicing and TCP Handoff are sub projects of LVS implementing Layer-7 load balancing.

For more information and great configuration articles and related downloads visit Linux Virtual Server

This project is to build an enhanced firmware for Linksys WRT54GS (versions 1.0 to 3.0 with 32MB RAM/8MB Flash) and similar routers, offering multiple Internet service providers Load Balance and Server Load Balancing capabilities (Layer 2). The Server Load Balancer technology is from by http://www.linuxvirtualserver.org/.

There are a number of routers supported by EzPlanet running Linux that have a base WAN (Ethernet) port and one or more LAN Ethernet ports. These routers could also be configured to load balance two (or more) WAN (ADSL) Internet links.

The latest version Ezplanet V24.1 has suport for web based IPVS status.

More information can be found here

Information on Multihome Load Balance can be found here