Blogging about web services and cloud computing may not be as much fun as writing a commercial for a tasty treat, but today’s announcement that Solace has partnered with Layer7 Technologies reminded me of that same idea: two things that are independently useful, and even better together.

When companies extend to the web using either web services or cloud computing there is generally a division between the management of security and policies for these outward-facing services and the interactions with in-house application components that comprise the service. It’s not a new problem, but many firms are still struggling with the implementation of large projects, and throwing racks of servers and armies of people at the problem to achieve necessary performance, reliability and scalability.

Appliances have taken off in the last few years because they make some pretty complex problems do-it-yourself friendly. Layer 7 has built a successful business providing turnkey solutions for web services, SOA, and cloud security, while Solace has made enterprise service buses rack-and-run simple with unrivaled performance and scalability. Before today, each company had a compelling appliance solution for one side of the equation.

Now Solace and Layer 7 have connected the peanut butter with the chocolate, so to speak, to provide an easy-to-use and truly end-to-end platform for enabling communications between internal and external systems and users. Two great tastes that taste great together!

This architecture is applicable to many applications and industries, for example:

• Government — Secure, inter-agency communications are a top priority for the U.S. government and many others around the world. National security and emergency response both rely on connecting the dots between information sources in real time so people can identify and respond to risks and emergencies as they take shape.
• Retail banking — Automating consumer services with a secure, high-performance and reliable web experience is essential for modern retail banking.
• Telecommunications — Mobile services and highly personalized provisioning and billing are the future for telecommunications providers worldwide.

Each of these industries faces the same fundamental challenge of securely delivering real-time information and interactive services to their (often huge) audience. We may not have groovy headphones or cowboy hats, but just like in the videos, the feedback from our customers suggests we’re onto something with this combination.

The press headlines are (mostly) celebrating passage of the financial reform act. I even got a bulk email from the president saying it was a great day for the little guy and a bad day to be a banking special interest guy. Then he asked me for 5 bucks and encouraged me to send the note on to 6 other people in my town. Wait, did the president just send me a chain letter?

But the reality is that financial reform is not at the finish line; the practical side of reform hasn’t even glimpsed the starting line. Financial reform is an agreement that more regulation and consumer protections need to be in place, but does little to define what those regulations or protections might be. There are some broad stroke intents related to making derivatives trading more transparent and changes in capital requirements that are fairly straight forward. But what information regulators will require from banks or specifics on how risk management will change are not yet understood. We don’t even know which bodies will do the regulating, or what their mandates will be.

The key for the banks will be in becoming more nimble and adjusting as the rules and regulations change and morph over the next several years. About a decade ago, corporate agility was at the forefront of business requirements, primarily because business opportunities were changing so quickly that firms with inflexible systems were being left in the dust. More recently cutting costs for higher profits has been in vogue, but we will almost certainly see corporate agility (probably disguised as some new buzzword) move back up the priority charts as a result of the Dodd-Frank act. Little else of the specifics are known, aside from the assurance that, as with all new regulations, lawyers and consultants will benefit.

To find out more about how to be ready for financial reform, please send me $5 and forward this blog post to 6 people inside your firm

In fact, the handling of this crisis is giving us a glimpse of the future of safety in open water drilling, now that money is less of an object. Here are a couple of recent articles that caught my eye:

• BP oil spill update: sensors measure spill — BP has installed sensors near the wellsite to improve their ability to estimate how much oil is spilling into the gulf. Some of the harshest criticisms have been around their inability to accurately determine how bad the situation is at “ground zero”. Sensors will help them more accurately understand oil flows and estimate what percentage they are capturing, as well as what degree of response is needed for the remainder.
• ‘Gliding’ robots patrol Gulf oil spill — BP has deployed water drones to swim around the gulf and measure data such as temperature, salinity and organic materials. By providing a much more accurate and real-time picture of the oil plume, that data will aid the planning and execution of containment and cleanup efforts.

If you just consider the financial and political cost of a single event like this, it would surely be much cheaper to outfit the areas subject to offshore drilling with a permanent set of sensors that can report on ocean quality and contaminants minute by minute, instead of two months after the next spill. The nice thing about sensors is they work cheap — once deployed they continuously collect data without interruption. And that data can be mined for a wide range of uses beyond the petroleum industry, helping with hurricane predictions and general environmental study.

This is not at all far fetched. As the cost of sensors and networks have come down, there’s been an explosion in fixed and mobile sensors, and the systems that are used to process the data. Also, there are already large scale efforts to monitor the oceans for environmental changes over time.

Every sensor network drives two complimentary needs: to identify long term trends using after-the-fact data mining  and to identify situations of immediate concern by analyzing real-time event streams. Solace is currently deployed in large scale sensor networks monitoring seismic and temperature events as well as aiding in efforts to protect major US cities from security threats.

There are many more projects where sensors are generating massive amounts of data that’s fed into a federated information bus for distribution to real-time analytics and data warehouses. Applications related to RFID-based inventory tracking,  smart energy grids, video surveillance, transportation flow & scheduling, logistics tracking and more. Location is an important factor in every one of these use cases, because the “where” and the “who cares?” aspect of sensor exceptions are at least as important as the raw data captured. In the consumer space we’re just beginning to tap the most abundantly deployed sensors in the world: our GPS-enabled phones. There is a staggering amount of possibility sitting in your pocket right now that will continue to change how we live, work and play.

That this trend is a harbinger of the future world doesn’t exactly take a lot of vision, it’s already happening. When we look back in 15 years at the sensors in the physical world, it will be like taking inventory of the Internet today vs its fledgling days of 1995. People were aware of the Internet, mostly as an email gateway, but few could envision the scale of information growth and accessibility that would give rise so services like Google Earth, Slacker, Wikipedia and YouTube.

The gulf oil spill has put the government between a rock and a hard place when it comes to offshore drilling. From the public’s perspective, this is like another 3 mile island, yet they won’t volunteer to give up their cars or turn off their heaters in the winter, which means demand for oil will continue to increase. Proactive monitoring of the offshore drilling zones with ocean sensors may be the best approach to regulating the offshore drilling industry while providing increased protection for the areas affected.

So when it comes to keeping track of the behavior of their trading systems, looking in the rear view mirror doesn’t cut it. Firms need to be on top of any problems that affect their trading operations, whether they’re in the acquisition and flow of market data, the automated execution of trades, or the pre and post-trade risk management that keeps things in balance.

For Solace customers engaged in high-speed trading, one of our major advantages is the highly-granular real-time operational visibility we give system administrators. Our solution uses takes advantage of the parallel nature of hardware to provide real-time per-client statistics that software-based systems can’t without impacting performance, and many of which can’t be collected at all in multicast environments.

For example, at the TCP layer our solution reports round trip time, bytes sent and received, queue depths, number of retransmits, and number of packets received out of order. And at the messaging layer, we provide instantaneous and high water mark data about queue depths, messages transmitted and received per second, and messages discarded due to queue depth problems. This information can help administrators quickly and efficiently identify and address the root cause of a wide variety of situations regardless of whether they are due to IP network problems or client application problems. This is all good stuff for middleware operations teams, but we fully realize that the visibility we provide stops at the edges of the messaging bus. The nature of middleware is that it always plugs into some other set of applications that comprise the end-to-end operational picture.

One of the more popular end-to-end monitoring solutions available is Geneos from ITRS Group. In their own words, their Geneos platform lets financial institutions “Gain control of every aspect of your environment: servers, applications, network hardware and critical network interfaces, as well as application behaviours and workflow characteristics — in a single solution.”

We’ve partnered with ITRS to let mutual customers monitor the behavior and performance of their Solace-based messaging environment as a seamless part of their environment. Together, we developed a NetProbe polling agent that directs operational statistics and monitoring data to the ITRS Geneos Active Console. The addition and integration of ITRS software to the Solace platform gives middleware operations teams as-it-happens alerting, and some seriously robust statistics for capacity planning.

A big thanks to the banking customers that originally suggested we integrate with ITRS. Geneos as a monitoring platform has a lot going for it: it’s innovative, widely adopted and well regarded within its installed base. With ITRS in our corner, high-speed trading firms have a powerful weapon to avoid becoming trading roadkill.

• Performance — for high-volume, highly repetitive tasks, special-purpose chips such as FPGAs, network processors and GPU’s have consistently been shown to outperform software running on general purpose CPUs. For many use cases, especially relating to market data and trading, performance alone is enough of a justification to choose hardware.
• Simplicity — the turnkey nature of many appliances such as messaging middleware, ticker plants, monitoring tools and security enforcement is appealing to many firms. They are so much easier and less costly to procure, deploy and configure that firms can focus on what they do best instead of getting bogged down with implementation details.
• Low TCO — appliances can often do the work of many equivalent servers running software, which significantly reduces the overall cost of operating an application or infrastructure.

But let’s be honest: simplicity and low cost don’t grab headlines like performance. If you have the choice to read a story about “12% cost reduction” or one about “performance measured in nanoseconds“, which one are you going to read? This is the equivalent of a World Cup commentator trying to convince the casual fan that Ronaldo is a special soccer player because of his mid-field play, while all we want is to see him crush the ball into the top left corner. Performance is where the sex appeal is, and it’s where most appliances make their name.

With that background, I am pleased to welcome NovaSparks as a new Solace partner. NovaSparks makes a hardware ticker plant appliance that accelerates and simplifies the processing of data feeds, and a trade order book for distribution to a wide range of trading applications. Sure, like other kinds of appliances NovaSparks is easier than software ticket plants and features lower TCO, but it is their eye-popping performance that will get your attention.

Ticker plants usually plug into a messaging layer for distribution, so Solace and NovaSparks have defined an architecture for integrating our two products to help customers move one step closer to a uber-performant end-to-end solution. To extend my prior metaphor, connecting our two high-speed appliances is like England’s Defoe slicing the ball from the corner to a deftly redirected header from Rooney for a game winning gooooooooooaaaaaalllllll!!!

Craig talked about how the technology pioneered as low latency solutions in financial services often leads to a better world when applied to challenges like homeland security, smart grids and transportation scheduling.

Brett talked about hedgehogs, S-curves and space time relativity. Yes, really! Brett’s session was definitely the most imaginative and thought provoking presentation at the conference.

If you missed the session or the event you can view their slides below (just click on the title page images) and listen to the audio here. Craig’s presentation starts about five minutes into the audio file, and Brett’s section begins right around the 18:00 mark.

Trading firms are transitioning from CPUs to field-programmable gate arrays, or FPGAs, graphics processing units, or GPUs, and cell processors. Such chips cut out the need for sending information across the operating system, and have the ability to run numerous processes in parallel, or simultaneously. CPUs actually have faster raw clockspeeds but can primarily only do one thing at a time, which ultimately slows things down.

“You’re basically sacrificing raw clockspeed for parallelism,” said Mike Dunne, chief technology officer at Activ Financial, which uses FPGAs in providing market data. “That trade-off was worth making, even though the clock speed’s lower, because we take advantage of the parallelism, as well as the power and the fine-grain memory control. You can get a 10-times or even 24-times increase in your performance.”

Solace was mentioned in the article by recently announced customer Liquidnet.

“We were frankly pushing our legacy systems to their limits,” said Keith Piraino, head of enterprise architecture at Liquidnet, a block-trading venue that recently announced plans to transition from software to hardware-based middleware from Solace Systems.

“We evaluated 10 different vendors for a solution and learned during this process that things are moving more into hardware,” he added, citing lower latency, the consistency of that lower latency, and lower costs among the reasons why Liquidnet chose to undergo hardware acceleration.

It’s nice to see one of our customers, and our story, get some recognition in such a respected publication.

Redline’s solution is a hybrid of software and hardware that is typically installed within a server to accelerate the various functions performed within a ticker plant. These includes everything you’d expect from a ticker plant: feed handling, order and price book functionality, even last value cache. Redline is fast, too — all of this takes place in an average of 8 microseconds for up to 10 million market data updates per second, with the kind of consistency you expect from a hardware-accelerated solution.

Our appliance picks up where a feed handler leaves off, reliably moving market data from feed to application whether that application is on the same machine, over a network or across the world.

So what makes the combination of Solace and Redline a match made in vendor heaven? There are two key scenarios:

• Co-location algo in a box — This is a configuration where Redline’s ticker plant is installed in a server alongside Solace’s API running as an IPC. Solace’s shared memory IPC protocol is specifically designed to distribute data at millions of messages per second between cores in a multi-CPU, multi-core environment with just a few hundred nanoseconds of latency. Together, Solace and Redline give applications full messaging functionality for high volume feeds with under 10 microseconds of latency.
• Single box market data distribution — In this configuration, Redline’s ticker plant is installed directly in the Solace message router, allowing fast and efficient handoff from Redline to Solace over the router’s internal bus instead of over the network. This enables lowers latency than having two separate boxes, and simplifies configuration and deployments.

The latter use case is much more common. High frequency trading and co-location get all the headlines, but market data is used virtually everywhere, far beyond high-frequency algos.

Cost considerations lead most firms to receive data through a shared feed handler, and use messaging to distribute it to algo machines, other trading applications, human traders, back office systems and remote locations.  Solace’s messaging uses the most efficient means to get the data from the feed handler to where it is needed:

• using IPC if the application is running on another CPU or core of the same machine,
• using hardware filtering and unicast to send the LAN application only what is needed,
• or using hardware filtering and compression over a WAN.

The Solace appliance automatically adjusts for each recipient’s configuration without the nightmare of configuring and managing multicast on the LAN or bridging out of multicast for WAN connectivity. There are so many ways the distribution use case can save people time, effort and money.

We look forward to working with Redline to make this consolidation of components 2010′s must have low latency trading solution.

In Formula One racing, all cars must comply with a defined set of rules (the formula) and find ways to differentiate with a car, driver and support team that all follow the basic guidelines.

The same is true of automated trading systems. The formula is simple: each trading system consumes market data and produces orders. What happens in between is how each firm differentiates themselves. Just like in F1, anything that can be improved is up for evaluation when it comes to achieving fast, consistent performance.

Today we announced a partnership with Solarflare Communications to accelerate linux-based networking of high performance messaging. Let me take a minute to lay out what this partnership does for our customers.

It is well understood that in the software world trading performance is negatively impacted by “context switching”. This is an operating system issue, where the OS needs to manage state as control is passed from “kernel space” where the network drivers reside, to “user space” where the application runs. Each individual context switch is fast, but when you’re talking about hundreds of thousands of context switches per second the impact on CPU and latency adds up quickly.

The Solace Message Router doesn’t have this problem because there is no software or operating system in the execution path of processing market data, so overhead stays low and consistent even as rates increase. The applications that connect to Solace’s hardware run in software on operating systems, though, so publishers and receivers are still subject to the effects of context switching.

This is where Solarflare steps in. They provide a network interface card (NIC) called Solarstorm and a network driver called OpenOnload that use “kernel bypass” techniques to run the application and network driver together in user space and, well,  bypass the kernel. This allows the application side of the connection to process many more messages per second with lower and more consistent latency.

Most firms value consistent latency as much as they do raw speed, and that’s where the value of this partnership really shines. We tested a system featuring Solace and Solarflare running between 1 million and 5 million messages per second, and the average end-to-end latency ranged from 24 to 35 microseconds, while 99.9th percentile results ranged from 29 to 46. For those volumes, those are ridiculously tight latency bounds. You can check out all the results in the Solace/Solarflare white paper here on this site.

If  you’re a developer or architect who has fought with context switching for years kernel bypass may feel like cheating, but fortunately it’s completely within the rules.

If dialog in the low latency finance community had that concept, it appears that “consistent latency” would be moving up the charts. For the majority of 2009, the focus was just on making everything faster. Lower the latency, shave a microsecond here and there, and so on.  This year, it seems that people have figured out that the harder thing to achieve is predictable latency. If you’re going to turn decision making over to algos, you need to be confident that the information that feeds them is current.

There was an article today at A-Team’s site where Scott Kennedy of Thomson Reuters was making just this point. His opinion:

“The future of competitive differentiation will be about predictability of performance instead of about speed.”

We agree with Scott. Consistent performance is one of the key reasons why trading firms are turning to hardware. Operating systems lead to variability and delays since they can get brain-locked on task switching in ways that hardware just doesn’t. This leads to the the enemy of trading profits — latency spikes.

Sure it’s fun being fastest, but it’s much more gratifying to see the topic of conversation evolve to what actually makes a difference to the business: making money. Well, that and talking cats.