Equinix key trends in financial data infrastructure

© 2010 Equinix, Inc. | www.equinix.com

Trends in financial daTa infrasTrucTure

Equinix WhitEPaPEr


Bringing together network providers, asset managers, brokerages, exchanges and trading platforms as well as market data and analytics providers, network-neutral data centers can dramatically lower the number and cost of high-speed interconnections.

summary

No stranger to turbulent change, the financial services industry continues to face revenue and cost pressures as well as an increased need to operate globally. Exchanges have merged and relocated, new financial centers and trading venues have appeared, and the market share for incumbent players has declined. Established players have been required to adjust both strategically and operationally to survive in this evolving marketplace. As a result, many firms are rethinking their technological infrastructure and evaluating co-located services.

Demand for fast, reliable and cost-effective technology has grown significantly in the last decade. Reliability and throughput are more important than ever before, and firms that were once content to operate their own data centers or co-locate infrastructure with a financial extranet are evaluating their ability to maintain a competitive position at a manageable total cost of ownership.

At the same time, the number of options to consider when determining where an order may be best executed has increased. As a result, the number of required connection points has grown. In this environment, the ability of network-rich data centers to connect to the various execution venues has made their value more apparent. Bringing together network providers, asset managers, brokerages, exchanges and trading platforms as well as market data and analytics providers, these data centers can dramatically lower the number and cost of high-speed interconnections.

Although data centers owned by network providers currently provide a range of co-location services, they have not proven effective in supporting traffic outside their own network. Increasingly, network-neutral data centers are seen as an effective option for firms pursuing global markets across the evolving financial ecosystem.

SINGAPORE SYDNEY TOKYO ZURICHPARISHONG KONGFRANKFURT GENEVA LONDONCHICAGO NEW YORK

Equinix Global Service Delivery Platform

Clearing Firms

Broker Dealers

FIX Network Providers

Asset Managers

News Feeds

Exchanges

Market Data Providers

Data Analytics Providers

Financial Extranets

Carriers

Equinix Global Financial Ecosystem

http://www.equinix.com


equinix WhiTe PaPer 2


Trends in caPiTal markeTs

As 2009 drew to a close, the Aite Group drafted an impact note outlining what it saw as the most significant trends affecting capital markets in 2010. Among the drivers:

n Market fragmentation and increased competition, particularly from multilateral trading facilities (MTFs) in Europe and alternative trading systems in the Canadian and Asia-Pacific markets.

n Increasing commoditization of high-frequency trading, particularly as independent traders have opened their own trading desks.

n Developing a global presence and managing global risk. Aite sees the importance of having a global presence as “one of the key issues for sell-side and buy-side firms.”

n Demand for trading infrastructure innovation, including low latency, throughput and support for multi-asset class trading.

Taken together, these trends underscore the important role that technology vendors will play in supporting participants in the global financial ecosystem. At the same time, those vendors are evolving, and the roles they play are shifting to better serve asset managers, brokerages, exchanges and trading platforms.

The regulators are coming

Commoditizing high frequency

Risk manager

consolidation

Going global back on the plate

Risk management goes real-time

It’s about the data

BPO and

BPO Pick up

Market fragmentation

continues

Back to the basics for technology

Clearing the corner

Source: "Top 10 Trends in Capital Markets for 2010." Aite Group, January 2010





The groWTh of co-locaTion services

The changes in roles for technology vendors started a decade ago, when financial services firms began looking for opportunities to ensure capacity (headroom), improve connectivity, strengthen business resilience and manage costs.

Financial services firms typically evolve their use of data center vendors in four stages:

1) Building a data center adjacent to or in a networked building

2) Co-locating in a phone company data center

3) Co-locating in a financial extranet data center

4) Co-locating in a network-rich data center

In the first stage, the local phone company serves as the primary network. Order flow from other geographic regions is limited by a carrier’s connectivity, often affecting the growth of the client’s business, as well. Capacity to grow space, power and cooling is typically limited in these installations.

In the second stage, the number of carriers to choose from likely expands. Because carriers compete against one another across many industries and markets, they typically are unable to bring those networks into the data centers they operate. The absence of multiple carriers can affect latency.

“Order flow from other geographic regions can be limited by a carrier’s connectivity, often affecting the growth of the financial services firm’s business.”

STAGE 1

Local Phone Company

Savvis

STAGE 2 STAGE 3 STAGE 4

BT RadianzTNS

Abovenet

AT&T

ColtEU Networks

SavvisLocal Phone

Company

Level 3

Local Phone Company

Extranet

Local PhoneCompany

Local Phone Company

Data center in exchange-owned building

Co-locate in phonecompany data center

Co-locate in financialextranet data center

Co-locate in networkrich data center

• Local phone company is the primary network.

• Carrier connectivity limits order flow from other geographic regions.

• Limited space, power, cooling and colocation headroom.

• Carrier connectivity expands.

• Carrier is territorial and does not attract competing networks.

• Add marginal space, power, cooling, and headroom.

• Access thousands of new end points from extranet.

• Extranets are often more territorial than phone companies


• Access dozens of carriers and extranets.

• Leverage global market data and order flow sources.

• Add substantial space, power, cooling, and headroom.

Many firms start out in their own data centers, using the local phone company for connectivity.

Once firms run out of space, they move to the phone company data center to gain more headroom.





The degree to which headroom improves when co-locating in a phone company data center depends on the provider. Data center owners that have made a commitment to this part of the business typically include additional capacity in their plans, making it easier for hosted businesses to expand. By comparison, a data center owner that outfits a facility for a small number of clients is less likely to allow room for expansion.

Financial extranet data centers give participating firms an opportunity to access thousands of new end points from the extranet, improving latency. Firms can also bring extranets into their own data centers, although this does not improve latency to the same degree. As with telephone carriers, extranets often compete across industries and clients, potentially complicating the interconnections. As in stage two, headroom is marginally improved.

The most mature and effective co-location strategy employs a network-rich data center. In such installations, financial services firms are able to access dozens of carriers and extranets, leverage global market data and order flow sources and add substantial space, power, cooling and headroom.

The network-rich data center also helps manage overall cost and improve throughput, typically driving down latency through improved access and connectivity. Six-sigma uptime (99.999%) supports business resiliency, as well.

STAGE 1

Local Phone Company

Savvis

STAGE 2 STAGE 3 STAGE 4

BT RadianzTNS

Abovenet

AT&T

ColtEU Networks

SavvisLocal Phone

Company

Level 3

Local Phone Company

Extranet

Local PhoneCompany

Local Phone Company

Data center in exchange-owned building

Co-locate in phonecompany data center

Co-locate in financialextranet data center

Co-locate in networkrich data center

• Local phone company is the primary network.

• Carrier connectivity limits order flow from other geographic regions.

• Limited space, power, cooling and colocation headroom.

• Carrier connectivity expands.

• Carrier is territorial and does not attract competing networks.


• Access thousands of new end points from extranet.

• Extranets are often more territorial than phone companies


• Access dozens of carriers and extranets.

• Leverage global market data and order flow sources.

• Add substantial space, power, cooling, and headroom.

As firms grow, they need more space and more connectivity options, so they move to a financial extranet data center.

The firms that choose network-neutral data centers generally see the fastest growth and most headroom while also dramatically increasing the connectivity options for clients and simultaneously reducing connectivity costs.

“In network-rich data centers, financial services firms are able to access dozens of carriers and extranets, leverage global market data and order flow sources, and add substantial space, power, cooling and headroom.”





comParing disTribuTed and hub (co-locaTion) models

To manage the growing complexity of the financial services ecosystem, many firms have pursued a model in which specialist services are distributed across a wide variety of network connections. In this model exchanges, data vendors, clearing firms, financial extranets, broker-dealers and others have operated as nodes on a far-flung network.

This approach is effective in supporting specialization within a firm, and network utility and system value of a financial network has increased as users and providers are added to the distributed ecosystem. Unfortunately, the nature of this global marketplace means that the need for high-volume, reliable interconnections has increased exponentially.

The distributed model fails as the number of endpoints increases. Using Metcalfe’s Law, we see that the number of endpoints required to support a distributed model increases from 45 (10 users) to almost 5,000 (100 users) and nearly 50 million (10,000 users). Eventually, the cost of maintaining a specialized, distributed infrastructure overwhelms the benefits associated with that system.

The distributed model has also struggled to maintain the capacity required to support algorithmic trading, among other demands. The result: bandwidth requirements are increasing rapidly, and infrastructure upgrades are trending toward continuous.

Exchanges

Market Data Vendors

Clearing Firms

Carriers

FinancialExtranets

NewsFeeds

Asset Managers

Data Analytics

FIX NetworkProviders

BrokerDealers

10 End-points = 45 Connections 100 = 4950 Connections 10,000 = 49,995,000 Connections

Metcalfe’s Law:

Reed’s Law: 2N–N–1

N(N–1)2

Distributed model fails as bandwidth requirementsand/or number of end points increase

Metcalfe’s Law Utility and systemic value of the financial network increases as users and providers are added. The need for connections increases exponentially.

Reed’s Law Eventually, the costs and infrastructure needs overcome the economic benefits of the system.

ThroughputAlgorithmic trading is driving up network capacity needs, dramatically increasing bandwidth require-ments and demanding continuous infrastructure upgrades.

“The nature of this global marketplace means that the need for high-volume, reliable interconnections with strategic customers and partners has increased exponentially.”





By comparison, a centralized financial ecosystem leverages economies of scope and scale for all participants. By co-locating within a network-rich data center, exchanges, data vendors, clearing firms, financial extranets, broker-dealers and others can dramatically reduce the number of circuits required to support multiple counterparties.

This approach helps financial services reduce and better manage costs while also improving throughput. Participants in a network-rich data center pay for one data connection into the data center, and counterparty transactions take place within the data center.

In this configuration, high-speed local area network (LAN) connections replace individual wide-area network (WAN) circuits that the distributed model uses to reach counterparties. The reduction in the number of circuits reduces costs even if bandwidth required and the number of end points increase.

The centralized or hub model also improves throughput. The efficient, network-rich data center configuration reduces latency to its physical minimum. The shared internal connections support significant growth, so individual user bandwidth can increase without incurring any cost increases or penalties.

“The reduction in the number of circuits reduces costs even if bandwidth required and the number of end points increase.”

ExchangesMarket Data Vendors

Clearing Firms

Carriers

FinancialExtranets

NewsFeeds

Asset Managers

Data Analytics


BrokerDealers

EQUINIXNEW YORK

EQUINIXLONDON

100 Counter Parties = 200 Circuits (96% reduction) The Law of the Hub: HN

Centralized interconnection points leverageeconomies of scope and scale for all participants

Cost�n Pay for one network connection — into the datacenter.�n Counterparty connections happen inside the datacenter.�n LAN cross-connections replace individual WAN circuits

to counterparties.�n Dramatically reduce costs even as bandwidth and end

points increase.

Throughput�n Latency reduced to physical minimum, while bandwidth

can increase exponentially without cost penalties.�n 99.999% availability.





The value of a shared centralized financial ecosystem is also seen in availability. As noted earlier, 99.999% uptime is the standard for these configurations.

Unquestionably, the financial industry will continue to scale, and its technology must increasingly do so in new, more predictable and more effective ways. By migrating from a distributed architecture to one that capitalizes on the “law of the hub,” financial services firms will be able to:

n Use dense interconnection points to reduce costs for circuits;

n Leverage economies of scale for all participants in the centralized network;

n Simplify connection needs by housing participants in a limited number of geographic locations; and

n Simplify network management by unifying connection standards in an environment in which throughput and uptime are critical measures.

Exchanges

Market Data Vendors

Clearing Firms

Carriers

FinancialExtranets

NewsFeeds

Asset Managers

Data Analytics


BrokerDealers

10 End-points = 45 Connections 100 = 4950 Connections 10,000 = 49,995,000 Connections

Metcalfe’s Law:

Reed’s Law: 2N–N–1

N(N–1)2

Distributed model fails as bandwidth requirementsand/or number of end points increase

ExchangesMarket Data Vendors

Clearing Firms

Carriers

FinancialExtranets

NewsFeeds

Asset Managers

Data Analytics


BrokerDealers

EQUINIXNEW YORK

EQUINIXLONDON

100 Counter Parties = 200 Circuits (96% reduction) The Law of the Hub: HN

Centralized interconnection points leverageeconomies of scope and scale for all participants

Dense interconnection points reduce costs for circuits.Economies of scale available for all the participants in the network.

Limited geographic locations simplify connection needs.Bandwidth can increase exponentially as needed.

“LAw OF ThE hUb”





rESiLiEnCECOnnECtiVitY

COSt COntrOLCaPaCitY

equinix co-locaTion services

Over the last decade, Equinix has developed a comprehensive network-rich data center solution built around four core principles:

n Connectivity

n Capacity

n Cost control

n Business resiliency

Proximity to the financial services client base is a key component of the Equinix data center business model. Each data center includes the clearing firms, exchanges, broker-dealers, market data providers, network providers, data analytics providers, asset managers, financial extranets and carriers in a given geographic space. Strategic locations in key global market centers, combined with in-building cross-connects, provide low latency and high throughput.

Execution venues typically place matching engines or access nodes in Equinix data centers, improving throughput. Internal to the data center, local cross-connects support high-speed data transfer (100 Mbps, 1 Gbps or 10 Gbps) at fixed monthly rates, proving participants with both capacity and cost stability.

Equinix is “carrier-neutral,” and it offers the use of networks and financial extranets that are critical to the success of its clients. Even in the firm’s most established data centers, up to 80% of the cross-connects go to carriers, making it critical to support efficient data transfer outside as well as within Equinix facilities.

Equinix maintains robust, diverse inbound and outbound connectivity options. Broad carrier choices give Equinix clients an extensive mix of high-bandwidth, low-latency network offerings, ensuring competitive pricing.

Equinix also maintains headroom for its clients to grow. Focusing on physical space, power and cooling, the firm’s goal is to operate strategically-located data centers and expand facilities and campuses to ensure Equinix’s ability to meet the needs of new clients and expansion requirements for existing clients.

This considered approach helps Equinix clients manage their costs, particularly when

compared against the private data center model. With Equinix, firms move

to LAN-based connections to accommodate demands for

higher bandwidths. This approach can dramatically reduce connectivity costs and

lower the trajectory for technology investments,

improving margins.

Finally, Equinix helps its clients maintain or improve business resiliency. The landscape for

market centers has changed repeatedly in the last decade, and market fragmentation and related regional competition is expected to increase.

In addition to a data center architecture that offers significant uptime, low latency and cost predictability, Equinix offers its clients flexibility and broad connectivity options that will outlast the changes, some not yet predicted, in the financial services marketplace.





equinix: enabling The global financial markeTPlace

For buy-side firms, Equinix data centers provide access to a broad range of trading, risk and compliance vendors. Each data center offers a resilient infrastructure proven capable of meeting the most demanding requirements. Equinix provides onsite connectivity to a broad range of liquidity pools, pricing and data vendors, trading systems and brokers. Equinix also helps buy-side firms reduce connectivity costs, simplify data center management, and maintain the ability to scale with ample headroom.

For sell-side firms, Equinix data centers offer low-latency connectivity to key execution venues and data sources. Each facility provides onsite access to regional, national and global carriers, simplifying interaction and reducing connectivity costs. Data centers also link members of the local financial ecosystem, improving access to potential customers and partners. Equinix provides sell-side firms with a broad range of choices in low-latency data feeds. The data centers support high-volume market data feeds at a manageable (and predictable) cost, even as volumes fluctuate.

For execution venues, Equinix data centers co-locate matching engines and access nodes, providing onsite carrier access and the headroom needed to accommodate a growing number of members and liquidity providers. Data centers also ensure proximity to market data distributors as well as access to members of the local financial ecosystem, increasing opportunities for order flow.

As exchanges merge and relocate and new financial centers and trading venues emerge, Equinix provides execution venues, buy-side and sell-side firms with the foresight and extensive connectivity needed to cost-effectively support their businesses in an uncertain environment. In choosing Equinix, members of the financial services ecosystem can be sure that their data center decisions will prove both cost-effective and operationally sound over the long term.

Latency Monitoring

Portfolio Management

Risk Analytics

Compliance

Pre-trade Analytics

Signal Generation

Trading Algorithms

Feed

Han

dler

s

Execution Algorithms

FIX

Gat

eway

s an

d C

onne

ctiv

ity

FIX

Gat

eway

s an

d C

onne

ctiv

ity

Smart Order Routing

Post Trade Settlement and

Processing

Post Trade Compliance

and Reporting

Order Management

Systems

Execution Management

Systems

Order Management

Systems

Exch

ange

Gat

eway

s

Real-time Market Data Feeds

Matching Engines

Matching Engines

Matching Engines

Matching Engines

Buy Side Sell Side Execution Venues





abouT equinix in The global financial markeTs

Equinix provides a neutral meeting place for the world’s leading financial market participants including trading venues, buy and sell side firms, market data providers, technology providers and financial networks; that locate servers and infrastructure within Equinix data centers in order to support highly reliable, low latency connectivity for a broad range of market participants. Learn more at: http://financial.equinix.com.

This white paper contains forward-looking statements that involve risks and uncertainties. Actual results may differ materially from expectations discussed in such forward-looking statements. Factors that might cause such differences include, but are not limited to, the challenges of acquiring, operating and constructing IBX centers and developing, deploying and delivering Equinix services; unanticipated costs or difficulties relating to the integration of companies we have acquired or will acquire into Equinix; a failure to receive significant revenue from customers in recently built out or acquired data centers; failure to complete any financing arrangements contemplated from time to time; competition from existing and new competitors; the ability to generate sufficient cash flow or otherwise obtain funds to repay new or outstanding indebtedness; the loss or decline in business from our key customers; and other risks described from time to time in Equinix’s filings with the Securities and Exchange Commission. In particular, see Equinix’s recent quarterly and annual reports filed with the Securities and Exchange Commission, copies of which are available upon request from Equinix. Equinix does not assume any obligation to update the forward-looking information contained in this press release.

Equinix and IBX are registered trademarks of Equinix, Inc. International Business Exchange is a trademark of Equinix, Inc.


http://financial.equinix.com

Economy & Finance

Equinix key trends in financial data infrastructure