Posted by (0) Comment
Data Banks of the Post-PC Era
by Chris Connor
Over the past year, makers of PC disk drives such as Western Digital {WDC}, Maxtor {MXTR}, Quantum Hard Disk Drive {HDD}, and even Seagate {SEG} have experienced difficult times. All four of these PC disk drive makers have witnessed lackluster revenues and stock prices. This trend will likely continue as PCs continue to decline in price. However, other areas of storage do not look so dismal. In fact, the demise of the PC is creating substantial growth opportunities in storage due to the overwhelming use of the Internet. All of the data that traverses the Internet has to be stored somewhere. Strong demand for non-PC storage also gets a boost from the increasing use of such handheld devices as Palm Pilots and cell phones. This strong demand for non-PC storage will likely raise the fortunes of several key companies in several key areas of storage.
At first glance, it would appear that enterprise storage behemoth EMC {EMC} stands to benefit the most from the vast amount of data transmission that the Internet is facilitating. However, there are other companies in other areas of storage that could benefit as much as, if not more than, EMC in the post-PC era. Are these other storage companies unknown diamonds in the rough selling at inexpensive valuations? Not really. Wall Street has already started its love affair with such non-PC storage stocks as Brocade {BRCD}, Gadzoox {ZOOX}, Network Appliance {NTAP}, and Sandisk {SNDK}. This attraction, however, unlike those based on a general affection for Internet stocks, is founded on some extremely solid fundamentals that are likely to continue to drive these stocks higher. Basic exigencies fueling this growth include the aforementioned need to store Internet data, the need to manage all of this data more efficiently, the ability to store MP3 songs in MP3 players, and the ability to store data in handheld computers.
Companies like EMC, Network Appliance, and Gadzoox should benefit the most from the need to store Internet data. Investors should take note that these companies focus on three different architectures, even though EMC sells products in the other two storage architectures. EMC is the worldwide leader in the enterprise storage system market. EMC’s primary products store, manage, retrieve, back up and share information from all the major computing platforms such as mainframe, Windows NT, UNIX, and Linux throughout the corporate enterprise of customers, suppliers, and employees. Network Appliance is the leading supplier of network attached data storage devices with a market share of close to 42 percent. Network Appliance’s products reduce the retrieval time and increase the accessibility of data stored on a network. According to the Internet Research Group, a market research firm, Network Appliance has the leading market share in Internet caching. Cache products bring frequently accessed data closer to the end user, thereby reducing waiting times for user access to web sites. Gadzoox, on the other hand, is a leader of a server-to-storage type architecture called the Storage Area Network, or SAN. SAN has received an amazing amount of investor interest. A SAN is basically a storage version of a LAN or Local Area Network connected by Fibre Channel. Fibre Channel is a high-speed storage networking technology that serves as the backbone for SANs. SAN represents the best performing networking storage technology today, as it eliminates bandwidth bottlenecks that were commonplace with older storage to server systems like SCSI. Currently, no other storage architecture can claim the level of speed that SANs are capable of delivering.
Companies such as Legato {LGTO} and Veritas {VRTS}, as well as EMC, provide network storage management software to protect data, ensure the availability of that data whenever it is needed, and manage the data efficiently. The beauty of network storage management software is that it is not limited to one storage architecture. Network storage management software supports all three of the storage architectures previously mentioned - enterprise, network attached storage, and SAN.
Flash memory takes advantage of the Internet as well, but not as a means of storing the abundance of data that accumulates on the Internet. Flash memory capitalizes on the explosive growth industry of portable devices such as Palm Pilots, wireless phones, and MP3 players. In addition, flash memory can be used in digital cameras and then removed to transport the images to the Internet to send to friends, relatives, or business associates, etc. Portable devices use flash memory products because of their small size and the ability to remove the memory whenever necessary. Flash memory is also non-volatile, which means that flash memory requires no power to store data. Companies primarily focused on flash memory products include industry leader Sandisk, Silicon Storage {SSTI} and M-Systems {FLSH}.
Posted by (0) Comment
Storage Area Networks: Are They Hyped or for Real?
by Chris Connor
Storage area networks (SANs) are basically storage versions of LANs or local area networks connected by fibre channel. Fibre channel is a high-speed storage networking technology that serves as the backbone for SANs. A SAN allows all storage devices to be available to all servers on a LAN.
SAN represents the best performing networking storage technology today, as it eliminates bandwidth bottlenecks that were commonplace with older storage to server systems like SCSI. Currently, no other storage architecture can claim the level of speed that SANs are capable of delivering, making it the storage architecture of choice for IT managers.
The SAN storage architecture is still in its infancy, but it has already found a killer application in the short term with the back up of data. Due to SAN’s broadband capabilities, it takes backup traffic off a company’s primary or non-storage network which enables a company to backup its data during business hours. Computer professionals recommend that companies have two to three backups of all of their files to be prepared should some unforeseen disaster occur. SAN also offers the advantage of distance. In a SAN, storage devices can be placed up to 500 meters apart by using fiber optic cable. The ability to separate storage devices by this much distance can minimize a company’s exposure to damages from a single catastrophic event. In other words, by separating its storage devices by significant distances a company stands a greater chance of losing only one storage device instead of several should a disaster occur.
The first article in this series mentioned Gadzoox in the SAN area, but there are several other significant SAN players such as Brocade {BRCD}, Vixel {VIXL}, Emulex {EMLX}, and Qlogic {QLGC}. Qlogic and Emulex focuses more on fibre channel than on the actual SANs - which means that it concentrates on the broadband aspects of storage technology. Brocade, the SAN stock with the largest market cap by far, also focuses largely on fibre channel by providing fibre channel fabric solutions to the SAN backbone. Gadzoox has become the leading SAN solution provider with a 66 percent market share in the SAN hub and switch segment and over a 98 percent market share in the SAN managed hub market. Vixel is a major industry player that just went public in September.
For all their potential, SANs are not without their problems. According to John William Toigo’s The Holy Grail of Storage Management, IT professionals must have extensive knowledge to manage SANs properly as SANs encompass both storage and networking technologies. This combination of network and storage technology makes SANs more difficult to manage than traditional storage systems. Normally, network managers do not purchase SANs. System administrators buy SANs, and they usually know little about LANs. Another weakness of SANs is that they are completely dependant on fibre channel. Fibre channel is largely a niche technology that is very expensive to integrate into a company’s larger network, which is usually Ethernet-based.
SANs are currently the best performing storage architecture - which makes them attractive to IT managers. The SAN architecture also looks to be well represented in the stock market with such high flying SAN stocks as Brocade, Emulex, and Gadzoox. However, investors may want to consider that SANs represent only a temporary solution for storage issues. The biggest advantage of SANs is the speed with which data can be retrieved. The SAN does not encumber other LAN functionality because the SAN is a network unto itself. However, what happens when a company’s regular computer network increases its bandwidth so much that it can support storage devices on that network without any significant performance loss?
Network Attached Storage Versus Storage Area Networks
by Chris Connor
Although some of the differences of Network Attached Storage (NAS) and Storage Area Networks (SANs) have been discussed several times in this series thus far, a full, blow by blow analysis of these two rapidly growing storage technologies is warranted. First, lets start with the basics. SANs are simply storage devices and servers connected together the way that computers are connected on Local Area Networks or LANs. A SAN uses fibre channel as its backbone while Ethernet usually serves as the backbone for LANs. On the other hand, Network Attached Storage connects to the network instead of having a separate network devoted solely to storage. An NAS appliance is not tied to a server like SANs are because it acts like a server itself - as an appliance linked to a network by an Ethernet card. This appliance usually consists of a group of hard disk drives, a processor, and its own operating system.
When comparing NAS and SANs, it is extremely important to the know the differences between Ethernet and fibre channel. Fibre channel is specifically targeted toward storage, while Ethernet is much more widely available as a network standard. Because it is a specialized technology, fibre channel makes SANs much harder to install and maintain than a simple network attached storage appliance. Only IT professionals with knowledge of both storage and networks should install and maintain SANs. However, that is not usually what happens. Most of the time, SANs are the responsibility of system administrators who are usually familiar with storage but not networks. On the other side, an NAS appliance simply acts like any other server on the network and is installed only with a web browser and installation wizard.
The one advantage that fibre channel currently gives SANs over NAS is performance. Because fibre channel is dedicated to storage, it currently offers higher speed than an NAS on an Ethernet network - since there are various other types of traffic on the network. The presence of these various types of traffic on a network is the reason that storage has been directly powered by a server instead of resting on a network the way an NAS appliance does. However, increasing bandwidth is changing all of that. Soon, Ethernet will likely pass fibre channel in performance as companies such as Cisco {CSCO} and Lucent {LU} are working on improving Ethernet speeds. Already, these two giants have created the fastest Ethernet so far - 10 gigabit Ethernet.
In terms of the sheer number of stocks with significant investment potential, the SAN market wins over NAS, hands down. The SAN market boasts several companies with investment potential while NAS has only one, Network Appliance {NTAP}. The SAN market consists of several components such as managed hubs, routers, adapters, and switches. There is usually one company that dominates the market for each SAN component - like Gadzoox {ZOOX} with managed hubs, Crossroads {CRDS} with storage routers, and Brocade {BRCD} with fibre channel switches. In addition, several SAN companies that make fibre channel adapters show investment potential - including Emulex {EMLX}, JNI {JNIC} and Qlogic {QLGC}.
However, some investors may prefer a single company that dominates its market the way that Network Appliance does in NAS. Network Appliance dominates NAS because of superior technology like its Write Anywhere File Layout (WAFL) system. WAFL enables efficient data storage of heterogeneous file formats from various environments such as Windows NT, Unix, and the Web. In fact, WAFL is the closest thing to true data sharing that NAS offers. The only way for true data sharing to become a reality would be if only one operating system were used - and that will likely never happen.
In a head to head battle, the simple NAS appliance should easily beat complex SANs over the long term as the superior storage standard. There will not be a need for specialized storage networks if the general network can easily support key storage applications such as the backing up of data. Are SANs doomed over the long term then? Not necessarily. Network Appliance and Brocade are spearheading the idea of a NAS-SAN combination to create a single, optimal storage standard. John William Toigo, in his book The Holy Grail of Data Storage Management, says “Such an integration would be a step toward improving the scalability of NAS storage, while setting the stage for enterprise-wide storage infrastructure management via a back end Storage Area Network.” Simply put, SANs will likely survive in some form over the long term. With that being said, investors do not have to limit themselves to NAS, with just Network Appliance as the market leader, while there are several leaders in SAN that should continue to grow at amazing paces in the short term. Pure SAN stocks like Brocade, Crossroads, and JNI have grown annual revenues in the 100 to 400 percent range. To sum everything up, NAS should emerge from a battle with SAN as winner, but SAN stocks still possess long term potential due to a possible NAS-SAN combination.
Posted by (0) Comment
DSL: The Baskin Robbins of Broadband
by Chris Connor
Just as Baskin Robbins purveys a multiplicity of ice cream flavors there are various entities that offer many versions of digital subscriber line (DSL) technology. For example, there is Asymmetric DSL (ADS), G.Lite, High Data Rate DSL (HDSL), ISDN DSL (IDSL), Symmetric DSL (SDSL), and very high data rate DSL (VDSL). ADSL is by far the most popular flavor of DSL because it caters to the needs of most Internet users. Instead of providing similar download and upload speeds like SDSL, ADSL provides substantially higher download speeds (1.5 to 9 Mbps) than upload speeds, because the majority of people download rather than upload. Upload speeds are not shabby either, since they range anywhere from 64 Kbps to 1.5 Mbps. G.Lite is basically a version of ADSL that goes through all the wiring of a house, thus limiting its speed because some of the bandwidth is consumed by noise. HDSL is a good replacement for T1, another broadband technology used by the phone companies, because HDSL does not require a repeater every 6000 ft to boost the strength of the signals like T1s do. IDSL is essentially DSL based on ISDN technology with data speeds of only 128 Kbps, but IDSL is capable of providing broadband access at significantly greater distances than ADSL. Although it is not widely deployed yet, VDSL offers higher speeds than all the other DSL technologies because the signals are transmitted only over short distances. The shorter the distance, the faster the access speed.
DSL is currently the best broadband technology offered by phone companies. DSL succeeds where the phone companies’ former attempts at providing broadband access (ISDN) failed. DSL offers both cheaper access and faster access speeds than ISDN because DSL bypasses the voice switch while ISDN has to pass through it. Not only does the voice switch slow down access speeds, but it must be upgraded before ISDN can even be made available to a customer. T1 lines are significantly better than ISDN, but T1s are limited to speeds of around 1.55 Mbps. With that being said, phone companies have a considerable interest in furthering the future of DSL because DSL is the phone companies’ answer to competition presented by cable modems. Most importantly, DSL leverages phone companies’ existing infrastructure while cable companies have to rebuild their networks to support two-way traffic.
Cable broadband access subscriptions are outpacing DSL subscriptions by a ratio of three-to-one because of two major weaknesses of DSL. First, DSL has a problem with distance. The farther away a customer is from a central office or CO (the location from which the telecom companies run the signals), the lower the connection speed to that customer. The major cut-off point at which performance really deteriorates is at 15,000 to18,000 feet. The second problem deals with the installation of digital loop carriers (DLCs) because DLCs block DSL signals. DLCs were added to some phone systems in response to the need for additional phone lines due primarily to Internet use. People wanted additional phone lines so that they could use the Internet while still being able to make and receive phone calls. These DLCs were used to add phone lines to households and businesses.
As a response to DSL’s current shortcomings, one DSL company may have come up with a solution that could propel DSL into the mainstream. Paradyne {PDYN}, once a part of AT&T {T}, recently introduced a new IDSL line card and router to its Hotwire system that bypasses DLCs and allows DSL providers to reach customers farther away than 15,000 feet without incurring a dramatic drop in performance. Paradyne is able to break the distance barrier for DSL with its patented Hotwire Multiple Virtual Line (MVL) technology. This breakthrough technology from Paradyne could be a major boon to DSL service providers because it allows them to spread their considerable costs over a larger installed base due to the additional customers that this technology makes it possible to reach. Lucent {LU} and several others are also working on curing DSL’s major weaknesses with products of their own.
Although a single company has not yet emerged to dominate the entire DSL equipment market, several companies have established strong positions in their respective areas. For instance, Copper Mountain {CMTN} leads in providing DSL equipment to businesses, while Alcatel {ALA} has the edge in providing DSL equipment to the consumer market. According to the Dell’Oro Group, Efficient Networks {EFNT} leads the DSL customer premises equipment (CPE) market since the company shipped 290,000 units of CPE products in the first quarter of 2000. As far as DSL deployment goes, an aggressive merger and acquisition program has propelled SBC Communications {SBC} to the top with over 300,000 DSL subscribers; this customer base comprises close to half of all DSL subscriptions. SBC consists of such powerhouse telecom companies as SBC Telecom, Ameritech, Southwestern Bell, PacBell, SNET, and Nevada Bell. Finally, Dataquest ranks GlobeSpan {GSPN} as the number one DSL chipset maker in terms of shipments. Investors should take note that Wall Street has placed the highest premium on GlobeSpan of all the major pure-play DSL stocks; Globespan registers a price to sales ratio of about 53 and a whopping market cap of about $5.3 billion. In other words, investors seem to think that GlobeSpan has the most potential of all the pure-play DSL stocks.
As opposed to cable, which has just one major ISP with At Home {ATHM}, DSL has both the RBOCs (Regional Bell Operating Companies) like SBC and Bell Atlantic {BEL}, and data-focused competitive local exchange carriers (CLECs) like Covad Communications {COVD}, North Point Communications {NPNT}, Rhythms NetConnections {RTHM}, and DSL.net {DSLN} to provide broadband access via DSL technologies to customers. This intense competition in the DSL market should further spur growth as DSL becomes available in more areas. Furthermore, there is a possibility that the DSL industry could work in cooperation with the cable industry since broadband is still in its infancy. The aforementioned CLECs, which are pure play DSL stocks, could offer DSL connectivity to markets outside a cable company’s reach. For example, At Home signed a multi-year strategic agreement with Rhythms to allow At Home can expand its overall broadband footprint. At Home needs Rhythms to expand its footprint because At Home’s cable partners do not own networks in most of the markets that Rhythms serves. The backbone for these DSL services will be supplied by At Home, while Rhythms’ network will serve as the “last mile solution” to the customers. Although copper phone lines should eventually be replaced by fiber optics, DSL will likely catch and surpass cable broadband access provision in terms of the number of subscriptions as the dynamic duo of the telephone giants and the pure-play DSL service companies drive impressive DSL growth over the next two to three years.
Posted by (0) Comment
The Need For Speed
by Chris Connor
With the explosion of on-line multimedia offerings such as movies, multiplayer computer games, and business applications, consumers have demanded faster access to these multimedia offerings than they have traditionally received via standard 56k modems - much faster in fact. Companies crave higher access speeds as well, because faster access will cut down on the cost of materials like CDs, paper, and video tapes. For example, the porn industry wants faster Internet connections so its customers can just download movies straight from a site, which eliminates the need for buying video tapes for each potential customer. The music industry also desires faster connection speeds, since most songs are a few megabytes in size and most modems today run at 56 kilobits per second. The answer to this need for speed is broadband A.K.A. “fat pipes”.
Broadband comes in several different forms, but the two big ones right now are cable modems and Digital Subscriber Lines (DSL), with fiber optics looming in the background as the biggest source of broadband for the future. Cable and DSL offer users connection speeds that are exponentially higher than 56k modems at around 1.5 megabits per second. In addition, both technologies stay connected to the Internet while a regular phone number must connect to the Internet each time a user wants to get on line and can be automatically disconnected during the worst possible times - like while chatting, playing games on-line, or downloading big files.
However, speed and the always-on feature are where the similarities between cable and DSL end. Cable access does not require a phone line, offers the Internet Service Provider (ISP) and high-speed connection from the same company, does not require long contracts, and it distributes bandwidth (the amount of data that can be transmitted in a fixed amount of time) that is shared by the cable users on a particular network. On the other hand, DSL offers a one-to-one connection between the customer and the ISP, and allows the customer to choose from several ISPs. With DSL, the data is transmitted over regular phone lines so DSL is better suited to handling two-way communications than cable - which was originally used for just receiving information.
What is the best way for investors to profit from the booming broadband industry? In light of the already intense competition among companies offering broadband access, investors would be wise to avoid that area and concentrate mostly on the equipment suppliers for DSL and cable access. Broadband equipment suppliers should continue to benefit from the immense growth in the industry, even as a plethora of broadband access providers emerges, because technology is much harder to duplicate than services.
The dominant cable ISP is At Home {ATHM}, but Terayon {TERN}, Broadcom {BRCM}, and Conexant {CNXT} offer better ways to invest in broadband over cable. Broadcom and Conexant make the chipsets for cable modems while Terayon makes the actual cable modem using its own proprietary technology called S-CDMA. Although Conexant is one of the biggest communications chip makers and is rapidly growing its revenues, Broadcom and Terayon appear to have greater futures in broadband access over cable due to tighter degrees of focus and superior cable technologies. Terayon uses its version of CDMA (primarily a wireless technology), S-CDMA (Synchronous-Code Division Multiple Access), to eliminate noise on even the oldest cable systems and to efficiently use bandwidth by differentiating the data using codes. Broadcom dominates the cable modem chip market because its chips have become the industry standard in a way similar to the way Intel’s chips have become the standard in the realm of PCs. In fact, Broadcom looks to be one of the broadband elite with its cable modem chip leadership and a growing presence in DSL, broadband wireless, and gigabit Ethernet (a local area networking (LAN) technology). Broadcom also recently started an optical networking division.
Although there are roughly three times more cable modem subscribers than DSL subscribers in North America, DSL does have a pronounced advantage over cable because of its one-to-one connection between the customer and the ISP. In other words, DSL customers do not have to share bandwidth with their neighbors. Leading the pure-play DSL pack are two companies that have been public less than one year: GlobeSpan {GSPN} and Copper Mountain Networks {CMNT}. Both Broadcom and Texas Instruments {TXN} are significantly larger companies that make DSL chipsets, but they are not pure DSL plays. GlobeSpan is also a leading developer of DSL chip sets and Copper Mountain Networks is a leading comprehensive provider of DSL solutions. Other major players include Adtran {ADTN}, Paradyne {PDYN}, Efficient Networks {EFNT}, Aware {AWRE} and Pairgain Technologies {PAIR}. (Pairgain is scheduled to merge with ADC Telecommunications {ADCT}). Bringing up the rear are DSL companies like Interspeed {ISPD}, Metalink {MTLK}, Tut Systems {TUTS}, and Orckit.
Investors interested in companies that are building infrastructure for broadband access might want to check out broadband network companies such as Global Crossing {GBLX}, 360 Networks {TSIX}, Metromedia Fiber {MFNX}, and Level 3 Communications {LVLT}. These companies sell their bandwidth primarily to ISPs and telecommunication carriers, so the broadband networks also benefit from the insatiable need for broadband without being saturated by competition. The leading barrier to entry for the broadband network market is considerable start-up costs, which is why none of these companies have generated any real earnings yet. However, some companies are leasing capacity before their networks are completed. Basically, the broadband network companies are willing to build their networks now at huge losses so they can reap immense profits in the future when these networks are completed. The biggest of these companies, Global Crossing, has a fiber-optic network that almost spans the entire globe - serving 24 countries and over 200 major cites. Global Crossing has essentially built an international fiber optic network that competitors will have a hard, if not impossible, time duplicating because of the time and cost required to deploy fiber under the oceans. Broadband networks must also be wary of building networks with old legacy equipment, because the network must be upgradeable for the future. Many companies will try to equal what Global Crossing has done, but very few will succeed any time soon.
Investors also may want to take a long look at a few of the major fiber optic makers. Fiber optics is a key enabler of broadband because regular phone lines simply can not compare to fiber optic lines when it comes to broadband levels of speed. With that being said, the fiber optic component industry is characterized by a mixture of large companies with sustainable competitive advantages like JDS Uniphase {JDSU} and Corning {GLW}, small companies with dynamic products that have just began to generate revenue like Avanex {AVNX} and Sycamore Networks {SCMR}, and one young private optic company that could not only revolutionize the broadband industry but the wireless industry as well. That company, Terabeam, looks to take the “fiber” out of fiber optics. Since building fiber optic networks all the way to a business or residence is prohibitively expensive, Terabeam looks to transmit high-bandwidth optical signals (starting at 1 gigabit per second) through the air. Not only does transmitting optical signals through the air bypass expensive infrastructure to each building, it does not require a spectrum license the way that microwaves do. Another key advantage that Terabeam’s optical signals have over radio frequency (RF) signals is that Terabeam’s technology is point-to-multipoint, while the overwhelming majority of RF technologies are point-to-point. With point-to-point technology, additional networks are required to tie all the points together as opposed to Terabeam’s technology, which will offer service to multiple customers through one hub. The biggest downside to Terabeam’s technology is that it can send an optical signal only 3 kilometers - so this limitation rules out using the technology in all rural and most residential areas. Nevertheless, Terabeam accelerates the practicality of an all-optical network because it offers a solution to the “last mile” (connection to the actual user) problem. All-optical networks benefit the whole fiber optic component industry because this industry supplies the building blocks for such networks.
A second key enabler of broadband that deserves attention from investors is digital signal processing (DSP). DSP is essentially enabling traditional copper phone lines to carry high-speed data because it compresses video, images, and voice into more transmittable sizes. DSP also boosts the capacity of Cable TV systems, which allows them to offer the hundreds of channels that have been derided as being excessive. According to Broadband Access Technologies by Albert Azzam and Niel Ransom, DSP “is even allowing power lines and home electrical wires in the wall to be used as high-speed data transmission lines”. The big four companies involved with DSP are Texas Instruments, Lucent Technologies {LU}, Motorola {MOT}, and Analog Devices {ADI}. However, Texas Instruments dominates the DSP market with close to a 50 percent market share.
Posted by (0) Comment
Posted by (0) Comment
Bull Market - A financial site dedicated to providing investors with sound, long-term investing advice.
Cabot Market Letter - Stock market investment newsletter which focuses on dynamic, fast growing companies with strong earnings, relative strength, and momentum.
Growth Report - The Growth Report is a weekly source of investment ideas and commentary that will come to
serve its readership well and far into the future.
RunStockRun - A growth stock newsletter for investors and traders, specializing in finding stock market winners.
Stellar Stock Report - Stock picks and recommendations for long term investors, in depth stock market overviews with precise market timing and buy and sell short recommendations.
The Strategic Investor - “Blue chip stock picks that will make you money and let you sleep. Discover the 5 blue chip stocks every investor must own now.”
Lemley Letter - Lemley Letter provides stock market analysis of interest to traders and value investors. Risk adverse investment philosophy focusing on the preservation of capital.