Ascend Communications? purchase of Cascade Communications last year has led to a flurry of innovation, following the departure of a raft of key executives to head networking startups.
According to one survey, no less than five companies with a Cascade connection were named as being among this year?s 25 most promising startups. And the majority have chosen Massachusetts for their home base to develop telecommunications switches, software and tools.
The collective aim of Argon, Sycamore, Redstone, Castle Networks, and Shasta Networks - which unlike its peers is based out of Silicon Valley - is to develop switching systems that are capable of moving data at trillions of bits per second and sending a mix of millions of data, video and voice packets to their destinations more quickly and accurately than today's systems.
But unlike earlier generations of firms that sold networking products and services directly to medium and large companies, the new wave of start-ups is targeting its offerings at existing telephone carriers - many of which are not set up to handle today's boom in digital data.
Tom Burkhardt, president and founder of Castle Networks in Westford, Massachusetts, said: "We are seeing a sea change in how telecommunications will be delivered in the future, and that means developing significantly better products quickly and at low prices to a relatively small group of customers, who will buy large quantities."
The company is building a switch to bridge the gap between voice and data networks.
Argon Networks, on the other hand, was formed 18 months ago, boasts $34 million in venture capital and employs 75 people. It is developing a high-speed switch router based on the Internet Protocol (IP), but which also supports asynchronous transfer mode (ATM).
GigaPacket Node is aimed at Internet Service Providers (ISPs) and telephone carriers and is intended to carry corporate data, supply Internet access services and cope with applications such as electronic commerce within an organisation.
Argon claims the product offers about 100 times the speed of existing transmission devices and provides Internet connections of telephone network-like reliability.
Michael Grady, Argon?s president, chief executive and co-founder, said "What you are seeing is start-ups tackling a piece of the telecommunications infrastructure they did not handle before because the major telephone carriers like AT&T used to do it all themselves."
He added: "So companies around here are working on bigger, more complex problems that require more capital and must be solved more quickly. The new entrants are going to use technology to their advantage to achieve cost and service differentiation."
Grady was a founding member of Stratus Computer, which is also now part of Ascend, but set up Argon with Steve Willis, one of Wellfleet?s founders.
But the executives agree that a telecoms startup must follow a rigorous timeline, if it is to succeed. It generally takes between 12-15 months to develop a product, six to nine months to market it to key customers before someone else jumps on the bandwaggon, and then another six to 12 months to reach profitability.
"Technology changes are moving so fast these days that if you need two years to develop your product, don't bother, because you'll miss the time-to-market window when you have little competition," said James Dolce, president of Redstone Communications.
Dolce previously co-founded Arris Networks, which was acquired by Cascade before its first anniversary. At Cascade, he acted as vice president and general manager of the remote access unit.
Redstone, meanwhile, was established in September, 1977, and raised $20 million from venture capital firms such as Matrix Partners, Northbridge Venture Partners and Battery Ventures a year later.
It introduced its Internet Edge RX edge switch routers in October, which are targeted at ISPs and Competitive Local Exchange Carriers (CLECs) and deliver intelligent traffic management capabilities through an ASIC-enabled architecture.
They perform packet classification, per flow queuing and scheduling at wire speed, enabling ISPs to deliver differentiated services and quality of service guarantees to their business customers.
Dolce said: "The age of the universal router is over. In its place, highly specialised, switch routers are emerging to address the scaleability, performance, reliability and service quality demands faced by service providers at the network's edge."
He continued: "While much industry effort is centred on solving the Internet backbone problem, little attention has been paid to the access problem. As business customers continue to embrace the Internet, service providers will seek new ways to deploy high performance and reliable access that scales to match the demands of an expanding customer base."
He added that network equipment suppliers will find themselves under pressure to deliver high-density, cost-effective packet switching devices that concentrate hundreds of customers onto Sonet links with high quality, wire-speed efficiency.
Another class of startups, however, is focusing on address and service translation gateways between voice and data networks. The gateways are intended to provide feature translation and protocol interworking between IP or ATM addresses.
This means that when using the Internet, customers can keep all of the voice calling services they are used to such as time-of-day routing, conference calling, messaging and feature dialing.
Castle Networks is building central office-based communications equipment that can act as a mediator between traditional circuit switched voice networks and packet switched data networks. The aim is to help integrate the public switched telephone network (PSTN) and the public switched data network (PSDN).
The firm?s Services Mediation Switch (SWS), which is due to ship in January, is the first in a family of products that will translate signaling and management information, and traffic between voice and data networks. The aim is to both reduce the cost of voice switching and to act as a go-between to the different application and networking protocols used by voice and data networks.
Tom Burkhardt, chairman, president and co-founder of Castle and a former Cascade executive, said: "Circuit switching is not dead, but the mainframe circuit switch must become a system of distributed switches and servers."
He continued that, while CLECs cannot go after an incumbents? market with a tool that is similar in function and cost to their rivals, "a Castle box can enable them to expand to a new city at one-twentieth the cost - translating to a 20-city risk versus a one-city risk for the same investment - dramatically lowering the cost of being wrong."
Castle Networks was set up in October, 1997 in Westford, Massachusetts and has been funded by $18 million in venture capital.
But the latest high-end networking player to come out of Western Massachusetts, Sycamore Networks, was established by the two former leaders of Cascade, Gururaj Deshpande and Dan Smith.
The firm, which employs 50 staff, has just won $20 million in two rounds of venture funding.
Deshpande, chairman of both Cascade and Sycamore, said the demand placed on public networks caused by data traffic would continue to increase to staggering rates, fueled by the Internet, corporate intranets and extranets.
"Carriers are in a race to provide needed bandwidth and simultaneously offer new services. The existing public network infrastructure is hindered by an inability to quickly adapt to new technologies and scale for capacity, constraining the rate of bandwidth growth required and the timely deployment of new services," he said.
Sycamore's goal is to exploit the physical resources of the existing infrastructure by adding optical networking capabilities to fibre optic networks to make them more intelligent. The aim here is to relieve current network congestion and lay a foundation for developing new services.
The most recent entrant into this arena is Shasta Networks, which was formed from a group of former Cisco heavyweights in March. It is venture-funded by two firms, Benchmark Capital and Institutional Venture Partners, which have provided $5.65 million in first-round funding, and the company only launched itself a few days ago.
Shasta is developing Internet and public network gateways, which are intended to enable ISPs, telephone companies and Fortune 100 companies to create highly tailored services for their users, particularly in the small and medium-size business sector.
Anthony Alles, Shasta?s 33-year old co-founder and president, said: "We believe that we are now on the cusp of a fairly big change in the business of service providers. In the past, access was expensive and the business case was primarily based on access alone. That's one of the reasons the network itself is fairly transparent today."
The system will aggregate users of specialised or value-added services at what Shasta calls the "subscriber edge," rather than aggregating content at the network level .
"Access prices have become very, very low as access pricing has become commoditised. If you are a service provider, you really have to think now about your next business model," he explained.
Also on board is Wu-Fu Chen, founder of Cascade and seven other communications companies. As Cascade?s vice president of engineering, Chen helped guide the firm through its initial public offering in 1994, but is now Shasta?s chief executive officer and chairman.
While many of these executives have been inspired by the success of Cascade to pursue their own goals, they realise they face a wealth of competition. A slew of well-funded firms, all with slightly different technology, is battling it out for the same niche.
But they all see that providing the technology to upgrade current carrier networks and future network build-outs represents a market too large for just Cisco, Ascend, Lucent and Nortel to exploit.
Argon?s Grady concluded: "It's not really a surprise that we see the same big picture at roughly the same time. What's unique about this carrier market is each has a slightly different view of the world. This market will be dominated by startups. The traditional enterprise players will have a hard time making the change to meet the carriers. Companies that most people have not heard of will become the Cisco's of tomorrow."
AI, VR and working on the morning commute - it's all to come
Development could pave the way for the next generation of rechargeable batteries
And Apple IS working on virtual reality headset
Indian bank falls victim to suspected cyber attack from North Korea's Lazarus Group