STAC Summit, 5 Jun 2017, NYC

Vendors with exhibit tables at the conference. (click here for the exhibitor list).

• Peter Lankford, Founder & Director, STAC

Peter will discuss the latest research and Council activities in compute-intensive workloads.

• Gaurav Chakravorty, Co-Founder, Qplum

Qplum is a hedge fund that successfully applies Deep Learning (DL) techniques to trading and investment. At the last New York STAC Summit, its co-founder and machine-learning expert, Gaurav Chakravorty, provided a few hints at the fund's approach to DL (see the video here). So we asked him back for a deeper dive. In this talk, Gaurav will detail the kinds of DL that Qplum has found useful, as well as some of the blind alleys they've gone down. He will then address some of the technical and operational aspects of trading using DL. What are the key bottlenecks in training and inference? Which software frameworks and which hardware platforms have proven most useful for those workloads? What does a deployment look like? What are the scaling challenges and the key drivers of cost? Also, how does DevOps work when a lot of the "Dev" is handled by machines?

• Luigi Mercone, Managing Director, Bank of New York Mellon
• Bill Fenick, Strategy and Marketing Director - Financial Services, Interxion
• Jeremy Eder, Sr. Software Engineer, Red Hat [Slides and video]
• Dave Payne, Vice President, Global Solutions Architecture, Elastifile [Slides and video]

For decades, business people have dreamt of a world where information technology was infinitely flexible: quickly and automatically scaling up and down in each location as business conditions demand. At last that dream seems to be nearing reality. The innovations keep coming, from virtualization, containers, and private cloud frameworks to software-defined everything and public clouds offering you-name-it-as-a service. While some areas of trading organizations can exploit this new agility paradigm, other areas face constraints, especially the front office. The need to minimize latency constrains both the location of resources and the tolerance for abstractions. Business logic needs to change frequently in response to market conditions, which requires a lot of data to be collected and analyzed with increasing sophistication. How are leading firms managing infrastructure scattered across their headquarters, dozens of colocation centers, and perhaps public clouds? Which workloads are the low hanging fruit? Which agility-enabling technologies are most relevant to high-performance financial applications? And what about the data? Liberating computation to roam free is great in theory, but not if the required data acts as a ball and chain. Our cross-functional panel will dig into these questions and more.

• Peter Lankford, Founder & Director, STAC

Peter will discuss the latest benchmark results involving big data workloads, including tick analytics, resource management in multi-tenant clusters, and optimizing Spark performance.

• Dan Eisner, Technology Executive (recently AQR, DE Shaw)
• Zahid Hussain, CEO, Vexata [Slides and video]
• Hollis Beall, Director of Performance Engineering, X-IO [Slides and video]
• Rob Peglar, Senior Vice President and CTO, Symbolic IO [Slides and video]

Analytic demands in finance are putting tremendous pressure on data infrastructures, whether its quants and AI programs searching for optimal trading and investment strategies or risk systems generating all the scenarios required by new regulations. At the same time, there’s momentous change on the supply side. The economics of flash and interconnects make it possible in theory to achieve leaps in both performance and performance per dollar (current shortages notwithstanding), while cloud providers continue to play a more significant role. With these trends clearly in sight, a new generation of innovators has begun to bring products to market that re-think how data is stored and accessed and how much that should cost. Their points of view have implications for how application architects think about data access. Our panelists will debate a number of alternative approaches.

• Peter Lankford, Founder & Director, STAC

Peter will discuss the latest research and Council activities related to low-latency/high-throughput workloads.

• Peter Lankford, Founder & Director, STAC

Organizations that are subject to timestamp-accuracy regulations such as Europe's MiFID 2 regulations (which apply to many US firms) and the CAT NMS plan must be able to demonstrate to regulators that they have used best practices to ensure compliance. Monitoring is part of this, but so is testing, which covers things that can't be monitored or that need to be prevented. The European regulators have in fact said that "relevant and proportionate testing" is required. But how does a firm persuade a regulator that its testing has followed best practices? And how should it relate its testing back to the ultimate goal of timestamp traceability? The STAC-TS Working Group has been hard at work to answer these questions with industry standards and the tools to support them. Peter will provide a brief review of what it has delivered and what is coming.

• Dr David Snowdon, CTO, Metamako

The accuracy with which we can measure the latency of a network-attached device is limited by the relative accuracy of the network timestamps used to calculate the latency. Likewise, measuring the absolute accuracy of hardware or software timestamps relative to a reference like UTC or NIST requires a device with higher absolute accuracy than that of the subject timestamps. So trading firms are keen to have a reference network device that is more accurate than anything else in their environment. At the last STAC Summit, David described the bootstrap problem this presents: how do you measure the accuracy of the reference device? That is, what’s the reference for the reference? He then outlined a methodology for measuring the absolute accuracy of network timestamps to the nanosecond. However, the only results he provided were for timestamps on 1GbE packets. Achieving this level of accuracy for 10GbE—and measuring it throughout the second between pulses of PPS discipline—is a more challenging class of problem. In this talk, David will describe such a methodology—vetted and enhanced by the STAC-TS Working Group—and reveal its first results.

• Dr. Matthew Grosvenor, Senior Systems Engineer, Exablaze

Across liquid markets, latency and jitter continue to shrink. This means that the differences in latency that can have a material effect on P&L continue to shrink as well. When it’s possible to get from tick to trade in a few hundred nanoseconds, firms start to care about every nanosecond in their information pathways. But one of the challenges in squeezing out nanoseconds is measuring them. For measurement to stay ahead of the measured, we must enter the world of picoseconds. Matthew spends a lot of time in this world, and according to him, it is a strange and interesting place. In this talk, he’ll take us all on a brief tour of the challenges, pitfalls and some unexpected findings along the way.

• Harry Foster, Chief Verification Scientist, Mentor Graphics

The number of capital markets firms developing FPGA-based solutions is growing. So is the pool of firms considering shifting at least some of their business logic to FPGA. As this trend continues, it’s helpful for technology managers to have as much context as possible about best practices in FPGA development, including both technical and human factors. Those who have not started down the FPGA path may wonder what it really entails. Those already neck deep in FPGA development may wonder how their experiences compare to those of their peers. As an organization that serves a broad swathe of the FPGA developer community, Mentor Graphics has a lot of hard data on these questions, in finance and many other industries. In this talk, Harry will share some of these findings. How are firms maximizing their FPGA development productivity? How do other FPGA shops look in terms of skillsets, team dynamics, technologies, and methodologies? Come to hear the answers and ask questions of your own.