Announce OVS Support

Pronto Systems announced it would be adding Open vSwitch (OVS) to its Open Switch Software Architecture (OSSA) suite.

Open vSwitch is a production quality, multilayer virtual switch licensed under the open source Apache 2.0 license.  It is designed to enable massive network automation through programmatic extension, while still supporting standard management interfaces and protocols.

Pronto has completed the integration of OVS 1.1.1 into OSSA 1.3 image. After booting up Pronto OSSA 1.3, users will be prompted,

Please choose which to start:
(Will choose default entry if no input in 5 seconds.)
[1] Pica8 XorPlus * default
[2] OpenFlow
[3] Open vSwitch
[4] System shell
[5] Change the default setting
Enter your choice (1-5):

The ported OVS code uses Pica8 Driver to configure and control the switch chips. While the Pica8 Driver is encapsulated with Pica8 API and released as a binary driver, the ported OVS stack on top of the driver is completely open sourced. Users can request the source code of OVS from Pronto Systems.

Why Does it Matter?

Open vSwitch (OVS) is one of the most popular open source implementation of OpenFlow 1.0.  OVS is the default virtual switch of Citrix Xenserver (project Boston). It is also integrated into OpenStack and Open Nebula Project.  The code has also been ported to a few ASIC based switches, but unfortunately none of them is open sourced.

By porting OVS to OSSA, we provide both the source and production-ready binary to researchers, developers, and administrators. This, we hope, will help the R&D community to continue enhancing the features and performance of the OVS switches.

Release Note

Here is the release note of the OVS port.

In Xorplus release 1.3, the Open vSwitch (OVS) is ported from openvswitch-7fb563b.tar.gz of master OVS branch.

This OVS module can be remotely controlled through ovsdb-server running on a different Linux platform (but users need to make sure the version of ovsdb-server is compatible to this OVS branch). When using ovsdb-server to control this OVS on Pronto switches, users need to configure the type of netdev and datapath as “pronto”.

For instance,

ovs-vsctl –db=tcp:10.10.53.53:6636 add-br br0

ovs-vsctl –db=tcp:10.10.53.53:6636 set bridge br0 datapath_type=pronto

ovs-vsctl –db=tcp:10.10.53.53:6636 add-port br0 ge-1/1/2 — set Interface ge-1/1/2 type=pronto

ovs-vsctl –db=tcp:10.10.53.53:6636 add-port br0 ge-1/1/3 — set Interface ge-1/1/3 type=pronto

Known issues of Pronto OVS in this release,

1. The following actions are not yet supported:

ODP_ACTION_ATTR_SET_NW_SRC,

ODP_ACTION_ATTR_SET_NW_DST,

ODP_ACTION_ATTR_SET_TP_SRC,

ODP_ACTION_ATTR_SET_TP_DST,

ODP_ACTION_ATTR_SET_TUNNEL,

ODP_ACTION_ATTR_SET_PRIORITY,

ODP_ACTION_ATTR_POP_PRIORITY,

ODP_ACTION_ATTR_DROP_SPOOFED_ARP.

2. Qos and queue for port config, and flow match of tunnel id are not implemented.

Source code distribution

The source code of the Pronto OVS is available through email request. Please send your request to support@pica8.com.

Advertisements

Notes On OpenFlow

OpenFlow has attracted significant attention in the past three months. Part of the reason is the announcement of ONF.  The OpenFlow Lab of Interop also helped pushing up the interest level. Well, inevitably, there also comes many doubts whether OpenFlow deserves such a high interest.

While OpenFlow is not our only focus in our Open Switching Software Architecture, we see it as a very potential technology that may change how we control and configure the network in the future. We would like to share our view on this technology.

Let’s be clear. Can OpenFlow solve all network problems?

Not likely, at least not in the short term.

OpenFlow is an control interface (or API) to program the switch data plane. It does not define new frame type (yet) or new way for switches to distribute the traffic. This means OpenFlow is still limited by whatever data plane capability that existing Ethernet can support. For example, with OpenFlow, Ethernet will still lose packets when congestion occurs.

OpenFlow is just a control protocol, what is new?

OpenFlow is in fact more like an API to program the switch data plane directly.

In the traditional network administration, users configure the protocols through CLI (or other configuration interfaces), then the protocols figure out how to control the data plane. With OpenFlow, administrators or software can directly control the data plane through OpenFlow interface.

Programming the data plane without protocol’s help used to be a no-no. So, what has changed?

Most of the L2/L3 switching technology was designed to respond to the changes of topology. Protocols are designed to handle addition, removal, and failure of nodes in the network. While this requirement is still quite important to many networks, some operators start to realize these protocols add tremendous complexity to the network management.

Take data center as an example. In today’s data centers, switches and cables are part of the infrastructure, which is well planned and laid out long before the actual servers are installed. In this type of network, the topology is fixed, ie. no need for STP to figure out the “best path” or OSPF to find the “shortest path”. All traffic and routes can be pre-determined.

In this case, if network administrators or software can directly program the data plane, either statically or dynamically, it will be much simpler than trying to distribute the traffic through ECMP or Trill from multiple vendors (or even different product lines from the same vendor).

So, what if failure occurs in the network? Well, since the topology is pre-determined, the error handling can be pre-programmed as well. There is really no need for individual switches to try to figure out the alternative route after the failure happens, especially in the case the failure is local to a handful or servers and most of the switches should not even care.

Practically, where can people use OpenFlow?

While still at R&D stage, OpenFlow is showing some interesting signs of solving real problems.

One particular potential area is to build a dynamic virtual network in data centers. With GRE or similar tunneling technology, OpenFlow can build a virtual network on top of fixed cables. This allows data center administrators to dynamically associate resources through this virtual network. Randy Bias of CloudScaling has published an insightful white paper on this architecture.

Another interesting area is to build an MPLS network through OpenFlow. Ericsson Research has build a Wiki to explain how this can be done.

An interesting coincidence in the two examples is they are both trying to use OpenFlow to program data traffic that tunnels through existing network. Well, it definitely helps to get through trial gap if OpenFlow does not require users to replace all existing equipments.

Finally, is OpenFlow a hype?

It does not feel like a hype yet.

While a group of people is over-excited about this new technology and another group of people is eager to dismiss it, most of the developers and network specialists are still investigating how OpenFlow is going to evolve.

Some people predicts it will only evolves into a niche. Well, compared to Ethernet, aren’t new switching technology, such as FCoE and Trill, all eventually become a niche?

Network Load Balance with OpenFlow

James Hamilton had an post on Network Load Balancing with OpenFlow. In this post, James re-iterates the vision of migrating networking to open and commoditized platforms, which we not only share but work hard to make it happen.

Open, multi-layer hardware and software stacks encourage innovation and rapidly drive down costs. The server world is clear evidence of what is possible when such an ecosystem emerges.

This post also gives an interesting example of using OpenFlow to load balance a multi-campus network.

… Essentially they distribute the load balancing functionality throughout the network. What’s unusual here is that the ideas could be tested and tried over a 9 campus, North American wide network with only 500 lines of code. With conventional network protocol stacks, this research work would have been impossible in that vendors don’t open up protocol stacks. And, even if they did, it would have been complex and very time consuming.

I find an interesting point in James’s conclusion – if it is not OpenFlow, it will be difficult (complex and time consuming) to try out the load balancing idea.

The question about the value of OpenFlow was often raised as an attack point to this new technology. People often quoted Professor Scott Shenker, a founding member of ONF, “[OpenFlow] doesn’t let you do anything you couldn’t do on a network before”. Some said OpenFlow was nothing new and did not bring value to the network world. This is only half of the truth. Take the load balance case as an example, that feature can be done on selected Cisco switches if you hire a network specialist for a couple of weeks, if not months, to try out various switches and configuration.

With the example of implementing a load balancing network with less than 500 lines of code, we see the potential of this OpenFlow protocol.

  1. New features can be done and tune through software, without involving manual configuration. This gives the hope a scalable network can be self managed and tuned by software.
  2. By allowing researchers and engineers to try out the new ideas on a production network, OpenFlow can roll out new network innovation much faster than traditional network environment.

This implies someday the system administrators might be able to use software, through OpenFlow protocol, to replace some tasks that can only be carried out by network specialists.  The system administrator can focus on more critical things like optimizing the efficiency of the whole data center or planning for the future growth than trying to trace down the packets on the wire.

Pronto announces 40GE product family

Pronto Systems Announces its 40 Gigabit Ethernet Products for High Performance Data Centers

Pronto 3980 and Pronto 3780 top-of-rack switches provide extensible software solution and complete 40GE interconnection from edge to the core.

Palo Alto, Calif., May 2, 2011 – Pronto Systems, Inc., a data center networking start-up, announced today its new Pronto 3980, a top-of-rack (TOR) 16-port 40Gbps Ethernet core switch, and Pronto 3920, a 10Gbps Ethernet edge with 40Gbps uplinks, for data centers to build high-performance  network interconnections.

Both Pronto 3980 and 3920 use Broadcom’s latest Trident chipset, providing non-blocking 1.28Tbps (Tera bits per second) of cross-sectional bandwidth with rich L2/L3 features. This new core-to-edge solution also incorporate Pronto’s Open Switch Software Architecture (OSSA) and provides software flexibility, manageability, and extensibility.   Pronto’s OSSA integrates Linux and open software stack Xorplus with life-time free license. This new architecture enables users to extend the protocol engines and to easily add management agent to the switches.

In addition to Xorplus and Openflow, Pronto also announced a new partner Switch++, which develops and supports CoreSwitch software.

“Switch++ is excited to work with Pronto Systems,” said Avi Gingold, Vice President Sales & Marketing of Switch++. “With advanced software architecture, modularity and flexibility, Switch++’s CoreSwitch Layer 2/2+ software will extend Pronto’s impressive capabilities and help reduce time to market for its deployment of High-Performance Switches.” CoreSwitch is available with Pronto’s entire family of Ethernet Switches.

“Pronto is committed to opening network platforms for innovation. We are excited to add Switch++ as a partner to provide more choices of software and services,” said James Liao, founder and CEO of Pronto Systems.

The list price of Pronto 3980 is $12,500, and Pronto 3920 is listed at $11,500. Both models will be shipping in July, 2011.

Pronto will display and demonstrate these two models with Xorplus at Interop booth #808. Switch++ will demonstrate the Switch++, CoreSwitch Layer 2+ software on these two models at Interop booth #2365.

Indigo Support for Pronto 3780

April 22, 2011 – Stanford released Indigo OpenFlow source support for Pronto 3780. This extends Indigo’s program to a 10GE platform.

From the OpenFlow Hub Forum:
http://forums.openflowhub.org/showthread.php?30-Indigo-release-for-Pronto-3780-48x10gig-platform

Announcing the release of indigo-2011.04.20 for the Pronto 3780. The Pronto 3780 is a 48 x 10-gig switch based on Broadcom 56840 silicon. The system architecture is very similar to the Pronto 3290 with a 2 GB CF card providing a large and flexible root file system.

This is a preliminary release with limited testing. Please read over the release notes before installing.

You can download the image from the OpenFlow Hub website:

http://www.openflowhub.org/display/Indigo/

If you have a chance to download and sanity check this release, please let me know.

Thanks much,
Dan

OSSA Deep Dive

What is OSSA (Open Switch Software Architecture)

OSSA is Open Switch Software Architecture which aims to provide an open software environment to increase the flexibility and manageability of data centers. OSSA includes three system layers, including boot loader, operating system, and software application.

At the boot loader layer, OSSA leverages uBoot, an open source boot loader with rich capabilities. By default, OSSA loads its pre-qualified Linux image from a local compact flash card. It is also possible to change the uBoot setting to bypass the local CF card and make the switch boot from remote DHCP servers for central management.

At the operating system layer, OSSA incorporates mainstream Linux kernels which support rich management tools. The kernel also supports native GCC compiler so users can develop or port their own management and monitoring agents to the switches. OSSA Linux uses the local Compact Flash card as the root file system, which allows users to install tools. Users can also install their own Debian or Redhat Linux to leverage the build-in tools in those distros.

On top of the operating system layer, OSSA supports two main software applications, Xorplus and Indigo. Xorplus is a GPLv2 open-source stack, supporting mainstream L2 and L3 protocols. For layer 2, Xorplus supports STP/RSTP/MSTP, LLDP, LACP, and VLAN. At layer 3, Xorplus supports OSPF, ECMP, RIP, and multicast protocols such as IGMP and PIM-SM. Additionally, Xorplus supports both IPv4 and IPv6. All Pronto switches come with life-time software license of Xorplus. Users can upgrade their software to the latest version of any available Xorplus release without additional charge. In case users need software support, Pronto offers different package for support service.

In addition to Xorplus, OSSA also supports Indigo software stack, an official OpenFlow stack developed by Stanford University. Users can choose to boot the switches into an OpenFlow device and manage them with any OpenFlow controller, such as controller of Big Switch Networks (BSN), Nicira, or NOX of Stanford.

How does OSSA help data centers?

The ultimate goal of OSSA is to enable switches to be managed and monitored like servers while still provide high-quality switching protocol stacks. By using the open source model on all layers, Pronto is to bring fast development and high quality software into the switching industry. The life-time free license offers customers peace-of-mind of free access to latest upgrade and software features.

Who supports the software?

Pronto provide various support service packages to its switches. The support comes with two levels, Gold Package and Silver Package. The Silver Package gets support through emails, while the Gold Package will have direct phone call support. The service price is per switch, regardless the features or switch speed.

How do I try out development?

Get a Pronto 3290. Bring up the switch and get to the Linux prompt. Write a “Hello World!”, compile it with GCC on the switch* (*ok, this is not supported yet, but we will add it by the end of June). Run it, There you go. Easy. If you are an experienced switch developer, you can use cross compiler on your server and FTP the binary to the switch.

What if all I need is a low-cost rich-feature switch?

All Pronto switches are pre-installed with Xorplus and will automatically boot into Xorplus CLI, which can be locally or remotely managed by any experienced networking IT. The syntax of the Xorplus CLI is Juniper-like, which is quite similar to Cisco CLI. You can find the details in the Xorplus manual.

Is Pronto OSSA open to partner with other software vendors?

Of course, we are an open company.

We welcome any forms of partnership and any partner that wants to contribute to the innovation of networking. Some switch protocol partners want to change business model to service end customers directly instead of selling the source code … welcome. Some network management software provider wants to partner on solution sale … definitely. Some soft router providers want to port their software to an accelerated platform to improve data plane performance … sure.

You can decide your own business model. We are completely open.

Pronto Announces Open Switch Software Architecture

Pronto Systems Announces Open Switch Software Architecture

Pronto Open Switch Software Architecture (OSSA) provides the ultimate software flexibility and manageability to the data center network.

Palo Alto, Calif., April 18, 2011 – Pronto Systems, Inc., a data center networking start-up, announced today its Open Switch Software Architecture (OSSA) which aimed to provide an open software environment to increase the flexibility and manageability of data centers.

OSSA leverages open source uBoot as the boot loader and Linux as its operating system. “By going with Linux, we provide an open environment for users and programmers to extend the manageability of the switches by installing and programming their own management modules. ” said Lin Du, CTO of the Pronto Systems, “Linux also provides the strong security and rich support tools that would help the data centers to manage and monitor the infrastructure.”

OSSA incorporates two major software engines, Xorplus and Indigo, into its protocol management. The Xorplus protocol engine provides rich L2/L3 protocol sets, including STP/RSTP/MSTP, LLDP, LACP, OSPF, RIP, PIM, and IGMP, with lifetime free license. Xorplus also supports both IPv4 and IPv6 for future-proof data center network.

The Indigo engine, an open source program developed by Stanford University, provides a new programming interface for users to manage the switches through OpenFlow controllers.

“We are excited to see OSSA emerge as an option for enterprise customers looking to leverage open standards, implement new features and deploy network virtualization in the enterprise with OpenFlow” said Guido Appenzeller, CEO of Big Switch Networks.

Besides the supported software, users can also use the GNU toolset to program, troubleshoot, and manage the switch with their own proprietary software. This provides the ultimate flexibility to the data center users.