Counterpunching: Cantor's Emergency Response

By Sean Gallagher  |  Posted 2001-10-29 Email Print this article Print
 
 
 
 
 
 
 

Problem: Connections with customer networks lost

Problem: Connections with customer networks lost

While eSpeed's engine for matching trades remained up and running, all connections to customers' private networks ran through optical lines at the World Trade Center.

Solution: Customers who had overseas offices connected to Cantor's London data center were rerouted across their own networks to London. ESpeed worked with customers to reconfigure their servers to point to London, and moved or expanded the permissions on customer accounts to connect to that site. For customers without overseas private networks, eSpeed worked to get them access over the Internet until the customers could get their high-speed connections hooked into the Rochelle Park facility.

Problem: Loss of customer routing information

Redirecting data from customers' private networks was close to impossible because the routing information was stored at Verizon's central office in the World Trade Center, and that office was destroyed.

Solution: From backups of the network control center applications restored at Rochelle Park, eSpeed was able to give Verizon information about the network's configuration. "We had to give them hints," says Noviello. Customers' private networks were finally pointed at Rochelle Park three weeks after the disaster.

Problem: Settling trades

Connections to banks were lost when eSpeed's data center in the World Trade Center was destroyed. That made it impossible to clear or settle trades.

Solution: ADP's ICI unit offered to handle eSpeed's and Cantor's settlement functions so that eSpeed could fulfill transactions. ESpeed sent the output of its trading system to ICI/ADP, and tested the connection Wednesday night with J.P. Morgan Chase & Co. and other banks.

Problem: Lack of space, equipment and people

The Rochelle Park facility was not expected to be a disaster-recovery location, and it lacked the phones and desktop computers to accommodate the more than 100 staffers relocated there after the disaster. On top of this, almost all of eSpeed's desktop support staff was killed in the disaster.

Solution: Compaq shipped more than 100 desktops and laptops to Rochelle Park on Tuesday. The equipment arrived Wednesday morning. Microsoft sent in desktop support people to help configure systems and get them up and running. Cisco's Network Supported Accounts team helped deploy a local network, including Catalyst 5500 switches, and brought in a telephone system based on Internet communications protocols.

Problem: Restoring data and applications

Nearly 100 servers needed to be loaded with software and account data in two days. The storage area network at Rochelle Park had only three tape units available to restore systems.

Solution: CTO Matt Claus borrowed an additional tape unit from a friend at another company to handle the load.



 
 
 
 
Sean Gallagher is editor of Ziff Davis Internet's enterprise verticals group. Previously, Gallagher was technology editor for Baseline, before joining Ziff Davis, he was editorial director of Fawcette Technical Publications' enterprise developer publications group, and the Labs managing editor of CMP's InformationWeek. A former naval officer and former systems integrator, Gallagher lives and works in Baltimore, Maryland.
 
 
 
 
 
 

Submit a Comment

Loading Comments...

Manage your Newsletters: Login   Register My Newsletters