Cisco Cisco Nexus 5010 Switch Dépliant

Pagina di 5
2   © 2011 Cisco and/or its affiliates. All rights reserved.
To establish competitive differentiation and profit in this market, firms are focusing 
on the following business imperatives:
•  Speed of execution
•  Instant access to market data
•  Ability to sustain peaks of activity
When peaks in traffic occur for a very short time (microseconds), they are called 
microbursts. A microburst can overwhelm a non-optimized network and cause 
transactions to be lost or require retransmission, which has significant negative 
impact on a firm’s financial performance. It is not unusual for a microburst to exceed 
a firm’s allocated network bandwidth or average link utilization by another degree of 
magnitude. Because of their transient nature, microbursts can go undetected if the 
monitoring tools are not granular enough.
The impact of microbursts on trading may be huge. What’s more, based on 
projected volume increases in the markets, the frequency of occurrences is 
increasing. One packet of information dropped because of a microburst that 
exceeds the network bandwidth can dramatically increase trading time, requiring 
retransmissions and possibly resulting in complete data loss and therefore 
transaction loss. Since an HFT algorithm performs thousands of trades per second, 
every retransmission can translate into significant revenue loss. The challenge is to 
minimize trading latency end-to-end while handling peaks without packet loss.
Solution Overview
Cisco’s Algo Speed High-Frequency Trading (HFT) solution is part of the 
comprehensive Cisco High-Performance Trading (HPT) solution portfolio and 
integrated architecture, built to address the trading value chain from the front office 
to the back office.
Firms establish and maintain their competitive edge by continually evolving their 
trading strategies and increasing the speed of trading. This stresses technology 
architectures that are not agile enough to be easily managed and upgraded. An 
agile architecture, such as the Cisco architecture underpinning Cisco HPT solutions, 
integrates the latest technologies from both network and application domains. It is 
modular, scalable, and manageable, providing a path to evolve each component with 
minimal disruption to the overall architecture.
The Cisco Algo Speed HFT solution delivers robust network connectivity without 
packet loss. The solution is tested and optimized for speed with the messaging 
middleware, the latency monitoring, and the server network adapters from Cisco’s rich 
ecosystem of partners. Built for HFT clients who have unique requirements for speed, 
reliability, agility, and manageability, the solution (Figure 1) includes the following:
•  Cisco Nexus® 5000, 3000 and 2000 Series Switches: Designed to meet the 
stringent, low-latency requirements of data centers for next-generation financial 
markets, these switches deliver cost-effective scalability with a transport that can 
navigate the transition to 10 Gigabit Ethernet and unified fabric, while maintaining 
operational continuity for an environment where system availability is assumed 
and maintenance windows are rare, if not totally nonexistent.
•  The Cisco Catalyst® 4900M Switch: An ideal solution for deployments that 
require high-performance services at wired-network speed, as well as high 
availability, and the modular flexibility to support Gigabit Ethernet, 10 Gigabit 
“The dominant factor in determining 
whether or not a high-speed network 
had a latency problem was how 
‘bursty’ the traffic was. In a gigabit 
local area network environment, for 
example, we saw that microbursts 
could result in large spikes of traffic 
that could induce upward of 500 
milliseconds of delays.” 
- Donal Byrne, CEO
Corvil
Corvil is the provider of the CorvilNet 
latency monitoring system, which 
detects microbursts and is used by 
many firms in the HFT business.