CROSS-REFERENCES TO RELATED APPLICATIONS

This application is related U.S. Provisional Patent Application No. 60/163,626 filed Nov. 4, 1999 which application is incorporated herein by this reference thereto.

BACKGROUND OF THE INVENTION

1. Field of the Invention

This invention relates to resources on computer networks, particularly the Internet, and more particularly to a file storage and retrieval system that is available worldwide via the Internet which additionally allows a direct transfer of Internet files to an Internet storage, retrieval, and sharing resource. The present invention acts in the manner of a “Internet hard disk” or “Internet hard drive” to provide online storage and retrieval resources for users.

2. Description of the Related Art

The Internet is the worldwide computer network making available a vast number of computer and information resources to institutions and individuals. A significant part of the Internet is the worldwide web that allows for web pages to be written in HTML and transmitted upon demand throughout the Internet. Recent developments have better established the use of XML (Extensible Markup Language) as a subset of SGML (Standard Generalized Markup Language, ISO standard 8879:1986). FTP (File Transfer Protocol) provides means by which files may be transferred over the Internet. All of these protocols are generally well known in the art, and collateral resources can easily be obtained to describe these further.

Additionally, portable programming systems such as Java®, JavaBeans, and JavaScript have been extensively developed with an anticipation of future portability across the vast network that is the Internet. Java®-related systems allow for object-oriented programming whereby objects or “beans” allow the passing of self-contained modules with associated processing methods that are used to act upon the accompanying data. Consequently, the “bean” can travel through a network and, under appropriate circumstances, have certain processes activated allowing manipulation of the information contained in the bean.

Advancements in Java®-related systems have given rise to the Enterprise JavaBean™ (EJB). The Enterprise JavaBean™ allows for clustering of servers such that the bean is given independence from specific servers on the system, yet can be activated or “instantiated” such that error recovery is easier, the system as a whole is more robust, and processing of the bean can be performed asynchronously so that all events do not have to happen at a pre-set time or serially/one after the other.

Enterprise JavaBeans™/EJBs allow serialization of beans. Such serialization allows the bean to be represented as a data stream of determined length. In essence, this is just a data file that is interpreted in the proper context, much the same as any electronic information file. Such serialization of the EJB allows it to be replicated and stored in case of catastrophic failure of a preferred server or the like.

If the server upon which the instantiated EJB dies, goes down, or fails, a previously replicated twin can be used to continue the process and allow for error recovery. More information about Enterprise JavaBeans™ technology can be found in the white paper, “Enterprise Javabeans™ Technology: Server Component Model for the Java™ Platform” by Anne Thomas, revised December 1998, prepared for Sun Microsystems, Inc. and published/made available by the Patricia Seybold Group of Boston, Mass.

Due to the nature of new technologies, terms such as “bean” or “instantiated” may seem unfamiliar to those new to the pertinent art. Reasons for this include the difficulty of communicating quickly new and complex subjects as well as the good-humored nature of those who intensely pursue the establishment of new technology, particularly software systems. Consequently, for Java®-related systems, a coffee theme is often present that indicates to those knowledgeable in the art the general subject matter of interest. While distinctions may be subtle in the art, they can be very important and serve the ends of those attempting to establish, share, and forward the technology.

Generally, home pages or other web pages are requested by the user through designation of the URL (Uniform Resource Locator). With the transmission to the user via TCP/IP protocol, the information present at the URL (and generally a file located somewhere on a computer) is transmitted to the user. The file may have links, or pointers, to other resources including images, graphics, audio or video streams, or other resources. Mark-up language is used on the Internet in an attempt to provide an open-ended structure by which information of any sort that can be stored electronically (or perhaps even otherwise) can be made available to an end user on demand. As such, the Internet is seen as a powerful tool making almost any information resource available to any computer or to any person using a computer.

Over the past several years, the personal computer has increased in power and capacity as commercial demand has driven the research and development of producers and vendors. It is now not uncommon to be able to easily find an Intel-manufactured 500 megahertz Pentium®-based system having well over 10 gigabytes of hard disk space, as well as 32-256 megabytes of RAM. As such, the power by which files may be received and acted upon by the local user through his or her PC has kept pace with the advances in technology.

However, there currently remain obstacles to universal access to an individual's own information stored on his or her computer. First of all, computers are very heavy. They are bulky. They generally weigh several kilograms and are not easily transportable. Lightweight laptop computers or the like generally do not have the same resources available to the user as a regular PC. Additionally, access to local area networks (LANs) is generally not available once the computer leaves the premises occupied by the LAN. Additionally, Internet access is often restricted by the use of a modem. Modems generally provide data transmission speeds on the order of 56 kilobits per second. This is approximately the same as 7 kilobytes per second. However, headers and other information are required to properly transmit information over the Internet and increase the effective size of files.

Even with the increased availability of broad band access to the Internet, it becomes an important feature of electronic information processing and the like in order to provide resident resources on the Internet. Such resources could include the sharing of files and the like in a manner that are easy to use and understand.

Due to these and other restrictions regarding data transport, transmission, and reception, a need has arisen for means by which files and other data may be available worldwide through the Internet and not tied to a local computer. The present invention addresses this demand by providing means by which files and other data may be stored on the Internet and made available worldwide through the Internet.

SUMMARY OF THE INVENTION

The present invention provides an “Internet hard drive” or “Internet hard disk” to and from which files may be stored and retrieved. Denominated commercially as “X:Drive,” the present invention allows users to store files of foreseeably any type on a resource available throughout the Internet. Once available to the Internet, the files stored on the user's X:Drive are available to the same extent as the Internet, namely worldwide.

Note should be made that the term “X:Drive” refers both to the system as a whole and to the individual space allocated to an individual user. Consequently, reference is sometimes made herein to the X:Drive system or to X:Drive to refer to the system as a whole. At other times, the term X:Drive indicates the user's individual X:Drive, or allocated space. The different uses are indicated by context.

In order to effect the Shared Internet Storage Resource of the present invention, a central or distributed storage facility is provided. First and foremost is the high-speed access storage facility where files are actually stored. Such individual storage areas may be allocated in individual limited allotments, or be left open-ended and limited only by the capacity of the physical devices responsible for storage. Metadata, that is data about the files stored on the network hard drives or other storage devices, is generated and stored in a separate database. The database of metadata (the metadatabase) and the network-attached storage facility may be linked by an internal network. It is possible for the database to be stored on the same network storage facility or device on which user files are also stored. System management may select whether or not to distribute or consolidate the database with the network storage.

Also attached to the internal network is a web server that serves to generate and transmit the information to the Internet, and ultimately the user. The web server files may pass through a load balancer and/or firewall before proceeding on to the Internet. The same is similarly true for information coming into the web server from the Internet.

XML may be used in combination with JavaScript or the like to provide two means by which the Shared Internet Storage Resource of the present invention may be achieved. The first is a JavaScript object which may be transmitted to a browser program running on the user's computer. Such browsers may include ones that are well known, including Netscape® Communicator and Microsoft® Internet Explorer. Alternatively, a stand-alone application may be installed and stored upon the user's computer. This stand-alone application serves to intermediate the user commands with the web server and ultimately the metadatabase in the Internet storage device.

As an additional enhancement, the user interface may be a client program that meshes seamlessly with standard user presentations in WYSIWYG (what you see is what you get) graphic user interfaces (GUIs). As such, a drive may be shown on the user's computer and may be denominated “x:” (or “y:” or “z:”, etc., depending upon user preferences). The user can then read from or write to the x:\ Shared Internet Storage Resource drive much in the same way as you would the local a:\ and c:\hard drive.

When the user shuts down his or her computer, information that is stored on the Shared Internet Storage Resource of the present invention remains on the Internet. The user can then access such information from another computer, another geographic location, or even give permission to share files on the Shared Internet Storage Resource with others. Password protection or other security protocols may be used to limit or discriminate access to the user's files.

The Shared Internet Storage Resource of the present invention allows for direct Internet-to-Internet file transfer to a user's allocated X:Drive file space in a process referred to as “Skip the Download” or “Save to My Xdrive.”

OBJECTS OF THE INVENTION

It is an object of the present invention to provide a Shared Internet Storage Resource on which users may store and retrieve files to make them available to themselves, or possibly others, throughout the Internet.

It is an additional object of the present invention to provide all manner of file access and control generally available to files local to the users for such Internet-stored files.

It is an additional object of the present invention to provide an easy-to-use and readily understood user interface through which files may be stored, retrieved, and manipulated on the Internet.

It is an additional object of the present invention to gather metadata regarding such files and to store such metadata in a database.

It is yet another object of the present invention to provide a plurality of means by which Internet-stored files may be manipulated and controlled.

It is yet another object of the present invention to provide a browser-based access to Internet-stored files.

It is yet another object of the present invention to provide stand-alone application access to Internet-stored files.

It is yet another object of the present invention to provide means by which Internet files may be stored on an Internet resource by a direct Internet-to-Internet transfer subject to the control of a remote or limited-resource user.

These and other objects and advantages of the present invention will be apparent from a review of the following specification and accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1

is a schematic view of the X:Drive system of the present invention. The different tier levels are shown, along with the marking indicia of a circle, triangle, square, and star/asterisk corresponding to the same indicia in FIG.

3

.

FIG. 2

is a schematic view of Java® library objects operating in the transactions or data exchanges occurring in the present invention.

FIG. 3

is a detailed flow diagram showing the operation of the present invention. Indicia including a circle, a triangle, a square, and a star/asterisk correspond to tier levels shown in FIG.

1

and indicate the level of operation of the steps shown in the flowchart of FIG.

3

.

FIG. 4

is a flowchart showing the operation of the XDFile Enterprise JavaBean™ (EJB) used in the present invention.

FIG. 5

is an overview of the Java® architecture used to effect transactions in the present invention.

FIG. 6

is an alternative schematic diagram of the Java® architecture shown in FIG.

5

.

FIG. 7

is a schematic and flowchart diagram showing the IO (input/output) for the database transactions of the present invention.

FIG. 8

is a schematic diagram of the data recovery process as effected by the FileIO component of the XDFile object used in the present invention.

FIG. 9

is a schematic depiction of failure recovery elements.

FIG. 10

is a schematic and flowchart diagram showing virus protection effected in the present invention.

FIG. 11

is a schematic and flowchart diagram of the Internet-to-resource transfer (“Skip the Download”/“Save to My Xdrive”) as set forth in the present invention.

FIG. 12

is a schematic and flowchart diagram of the client system used in the present invention.

FIG. 13

is a Windows™ desktop display showing both the client and web-browser applications.

FIG. 14

is a display of a web browser pointing to a user's X:Drive.

BRIEF DESCRIPTION OF THE APPENDICES

Appendix 1 is a listing of web site/server code use to achieve the present invention.

Appendix 2 is a listing of the code used on the client side to achieve the present invention in a Microsoft® Windows™ environment.

Appendix 3 is a listing of the JavaScript code used to achieve the present invention in a Sun Microsystems® Java® environment (including one on a browser).

DESCRIPTION OF THE PREFERRED EMBODIMENT(S)

The detailed description set forth below in connection with the appended drawings is intended as a description of presently-preferred embodiments of the invention and is not intended to represent the only forms in which the present invention may be constructed and/or utilized. The description sets forth the functions and the sequence of steps for constructing and operating the invention in connection with the illustrated embodiments. However, it is to be understood that the same or equivalent functions and sequences may be accomplished by, different embodiments that are also intended to be encompassed within the spirit and scope of the invention.

Appendices 1, 2, and 3 provide the source code for, respectively, the Web Site/Server Code of the X:Drive Shared Internet Storage Resource system of the present invention; the Windows Client Code; and the JavaScript Listings for the present invention. These Appendices are incorporated herein by this reference thereto as if set out in their entirety. It is contemplated that these Appendices provide a full, complete, and enabling disclosure to those of ordinary skill in the art or less by which the present invention may be achieved.

Additionally, the reference numbers used in conjunction with the figures are numbered such that the 100's place of the number indicates the number of the drawing figure. For example, the 600 series of reference numbers refers to

FIG. 6

, while the 200 series refers to elements shown in FIG.

2

.

The present invention provides a method by which an Internet hard disk or hard drive may be achieved in a manner similar to a hard disk or hard drive available locally to the individual on the local computer. Additionally, as Internet use becomes a more familiar and everyday event for people, the resources provided by the present invention may allow the actual use of the Internet hard drive or X:Drive set forth herein to act as such a resource with the files being called up for execution for programs available and processed either locally and/or over the Internet. In light of the foregoing, it can be seen that the present invention may act as a bridge or may pave the way towards a more inter-networked community for the use and processing of electronic information.

The virtual disk drive provided by the present invention may be selectively shared with others or kept entirely private. Additionally, and as set forth in more detail below, the use of a metadatabase provides quicker access and the ability to distribute the information regarding the legion of X:Drive accounts over a wide geographic area, enabling redundant preservation of user information by server clusters implementing Enterprise JavaBeans® (EJBs), or otherwise.

The Shared Internet Storage Resource, User Interface System, and Method set forth herein is generally referred to as “X:Drive.” Context reveals whether or not the term X:Drive is referring either to the system as a whole or the individual's own account.

The X:Drive system of the present invention uses network application practices and may rely upon Java® Enterprise JavaBeans™ (EJBs) to enable distributed and clustered computing and file management environment. Along with such Java®-based and network-oriented design, the X:Drive system of the present invention also contemplates the use of open programming standards such as XML and Web-DAV (Web-based Distributed Authoring and Versioning). The use of such technology is foreseen as providing wide support by the user community as well as speed and development, refinement, and polishing.

As shown in

FIG. 1

, the X:Drive system

100

has a multi-tiered, network-based application infrastructure. The multi-tiered nature of the system allows it to separate operations in an efficient manner. The network-based aspects of the X:Drive system allows it to disperse resources geographically as well as allow a high degree of communication between different aspects or facets of the system.

The X:Drive system may be considered enabling technology as a medium that is independent of the applications and uses to which it is applied. The X:Drive system is currently based on object-oriented principles with each application layer responsible for a discreet functionality or aspect of operation. Both hardware and software resources may then successfully experience heavy re-use with both scalability and flexibility inherently provided. While these advantageous aspects of the X:Drive system are achieved, as a multi-tiered system, X:Drive involves a higher cost of complexity and planning. Thus, those who would seek to wrongly copy the X:Drive system would do so without accruing the great expense in time and money necessary to achieve the present X:Drive system. They would ride on the backs of those who not only developed the system, but also those who got it to work right and in a commercially-reliable manner.

The use of tiers in the X:Drive system of the present invention is realized in both the network systems and the application systems involved in achieving X:Drive.

As shown in

FIG. 1

, a variety of tiers, or layers, are present between the client

102

and the ultimate data resources

104

. Between the client

102

and the data resources

104

, are one or more layers or tiers, accomplishing the following.

The client

102

may be coupled to a public network

106

(such as the Internet) that may include a DNS redirector

108

as well as a load balancer

110

. The public network

106

may then lead into a web server network

120

. The web server may then lead into an application network

122

, which in turn leads into an EJB (Enterprise JavaBeans™) network

124

. The EJB network

124

may lead into a transaction network

126

, which in turn leads into the data resources

104

.

The client

102

may be either a web- or browser-based application or an application resident on a Windows™ X system (the X indicating the version of Windows applicable, i.e., Windows® 95, Windows® 98, Windows® 2000, etc.). Requests generally originate from the client as the X:Drive system

100

is one that operates at the command of users directing the client program. Client requests may be made versus the Hypertext Transfer Protocol (HTTP) GET/POST function. In a preferred embodiment, the GET/POST operation may be augmented with Web-DAV extensions to the HTTP protocol. Commands are transmitted by the client

102

are sent to the DNS redirector

108

, which then isolate the request via a proxy server process. A proxy server process prevents a direct connection between the client

102

and the other downstream resources in the X:Drive system

100

. Such proxy serving prevents inadvertent or mischievous disruption of service by allowing only certain commands or information to be propagated through the X:Drive system

100

. This prevents mischievous users from disrupting the system as such rogue commands are intercepted by the proxy server and denied further propagation.

After the client command has passed through the DNS redirector/proxy server

108

, the request by the client

102

is then directed to the most appropriate facility. As the X:Drive system is scalable, facilities may be distributed geographically, even over the face of the globe. This allows, at the outset, more efficiencies to take place in the X:Drive system

100

of the present invention so that more users may be served more quickly and so that the advantageous features of the X:Drive system may be realized by the widest number of users in the quickest way possible.

Due to the construction and architecture of the X:Drive system

100

, a number of machines/servers running a number of different processes may be distributed over a wide area. Broad band or high-speed access as provided by Internet backbone or the like may allow the X:Drive system to be effectively carried out over the entire face of the planet. The scalability and flexibility of the present invention augments its utility. Such advantages are further advanced by efficient use of the resources so that greater and better service can be provided.

Upon receiving the request from the client

102

, the DNS redirector

108

transmits the requests on to a load balancer which may provide a second proxy process under HTTP protocol and transmit the request to the least-loaded and most-available web server on an internal, non-routable, or other server network

120

.

The web server network

120

may be non-routable and may comprise a number of individual machines or servers processing the HTTP or other requests from one or more load balancers

110

. Each of the web servers

140

in the network

120

may handle HTTP requests for static content, such as HTML and graphic files. The web servers may proxy all requests for dynamic content to a Java® application network

122

.

As used in the X:Drive system

100

of the present invention, the Java® application networks may be non-routable. The use of non-routable facilities in the X:Drive system

100

of the present invention indicates their operation in a local area network (LAN). However, between tiers, the individual networks themselves may be available such that a web server

140

in Illinois may pass requests for dynamic content to Java® application clusters

122

in Wisconsin.

Each Java® application cluster

122

may be composed of a number of Java® application servers

142

with each server

142

handling display functions necessary for user accounts, including the generation of XML, HTML, and other instructing displays for either browser or application clients

102

. If a Java® application cluster

122

receives a resource request from the web server tier

120

, the Java® application cluster

122

will pass the resource request onto the Enterprise JavaBean™ EJB network tier

124

.

As for the web server

120

and Java® application networks

122

, the EJB network

124

may also be non-routable and operate upon a LAN. The EJB network may be an EJB cluster having a number of EJB servers

144

. Each EJB cluster handles the business logic and resource access methods and protocols required for the resource requests and management. The EJB cluster (EJBC) caches memory of common resources such as the pooling of data connections and the like, as well as data objects. Resource access requests and transmissions are then passed out to the transaction network tier

126

, which may also be non-routable. The transaction network tier

126

has a transaction processor

146

which controls, operates, and guarantees access and transactions on different resources. These different resources are the ultimate data resources

104

that may include NFS (Network File Server) disk arrays

150

and databases

152

. The NFS disk arrays

150

may supply the actual storage capacity for the files of generally any size. The databases

152

comprise records of information regarding each of the files (metadata) stored by the NFS disk arrays

150

under the X:Drive system

100

.

By bifurcating the file information in databases

152

separate from the actual files themselves on the NFS disk arrays

150

, file information and user queries can be handled much more quickly as display components of the present invention are important to provide the user information regarding the status and availability of the files stored on the X:Drive system

100

. Consequently, although a user may have a hundred separate files in an X:Drive directory, he or she may be only interested in one. Consequently in order to provide the user the information necessary to make the decision as to which file to receive, move, rename, delete, or store, the use of the database provides a very quick and easy means by which such user requests can be satisfied. It is anticipated that the actual use of the file storage facilities on the NFS disk arrays

150

or the like may comprise only a part of the operations of the present invention. Having the ability to display, select, and determine file operations is one of the useful advantages provided by the X:Drive system

100

of the present invention.

Note should be taken of the non-numerical indicia present in FIG.

1

. Most notably, a circle is associated with the client

102

, a triangle with the Java® application cluster

122

, a square with the EJB network

124

, and a star/asterisk with the transaction network. These non numerical indicia correspond to those set forth in FIG.

3

. As different actions are performed at different tiers in the present invention, the non-numerical indicia provide an easy or visual means by which the operation of the different tiers can be indicated in FIG.

3

.

FIG. 2

shows a logic diagram in sequence structure for the Java® library objects used in the X:Drive system

100

of the present invention. Generally, throughout the description of the X:Drive system

100

of the present invention, the prefix XD indicates “X:Drive.” For example, in

FIG. 2

the steps/status indicators of XDError stands for X:Drive Error, and XDXML stands for X:Drive Extensible Markup Language. Likewise, the use of the term XDFile indicates X:Drive File as a Java® library object effecting and intermediating the file operations of the present invention.

In

FIG. 2

, the Java® system

200

allows operations to be performed on the metadatabase

202

and the operating system (OS) File System

204

. Additionally, the XDFile object

210

may activate or instantiate the Database.Search object

216

. The XDFile object

210

may be activated, or invoked, by the FileAction object

220

. The FileAction object

220

may also activate the Database.Search

216

and Database. BigSearch

222

objects. Operations of the Java® library objects in the system

200

as shown in

FIG. 2

may be contingent upon the SessionSecurity object

224

, which may instantiate or use the Database.Search object

216

and/or the Database.Transaction object

214

. The SessionSecurity object

224

may return a separate object

226

to the UserData object

230

. The Database object

236

may inherit or transmit from its Transaction

214

, Search

216

, and/or BigSearch

222

objects.

The information generated may then be transmitted to the Database

202

for meta-information and the OS File System

204

for the actual data. If an error is generated during the operation of the Java® library object system

200

, an XDError object

240

may serve to handle the error while a successful operation may be returned in the form of the XDXML object

242

. In the Java® library object system

200

of

FIG. 2

, the Database

202

may contain intelligence or programming for connection to SQL databases and the like. Options regarding the operations of the database

202

may be read from a configuration file. The Database object

236

may be able to connect multiple databases for redundancy in the case of repeated or redundantly archived information, or for functionality in order to connect to that database which responds most quickly to the requests and commands.

The Database object

236

determines which database operation to perform and/or to which database to send operations based on the type of request it receives. For example, transaction requests may demand a separate database from those of regular query and BigSearch

222

requests. In order to maintain more efficient operation, the Database object

236

generally sends session users to the same database whenever possible so that latency and database replication is not passed on to the user.

The Database.Transaction object

214

is able to handle larger SQL statements such as those that would cause a load on the database. The Database.Transaction object

214

may spawn children classes that handle the transaction logic in order for more efficient operation.

The Database.Search object

216

is designed to handle smaller SQL statements and has children classes for specific search types, such as those along anticipated and common fields or types of information.

The Database.BigSearch object

222

handles larger, non-transactional SQL statements such as those used for reports in system accounting, monitoring, or otherwise. Children classes of the Database.BigSearch object

222

would handle specific large searches such as those that might be implemented on a monthly or other periodic basis.

The FileIO object

212

inherits and overrides Java®'s data file object. The file object contains logic to engage multiple disks or resources for redundancy and/or functionality and contains the functionalities necessary to manipulate files on the OS File System

204

. The FileIO object

212

may react to the JMS (Java Messaging Service) events triggered by events on the disks of the OS File System

204

.

Alternatively, one or more monitoring objects may be used to gather pertinent status information regarding the OS File System

204

. When monitoring objects are used, the FileIO objects then query the common monitoring objects to determine the state of the system. In the present system, the monitoring object is denominated the Mount Point Status bean, or MPS bean,

534

(FIGS.

5

and

9

).

Additionally, disk level transactions are carried out by the FileIO object

212

. Under the management of the FileIO object

212

, user accounts are able to span or traverse several disks. The spanning of such several disks enables better recovery from failure should an error occur or system resources become unavailable in an unpredictable manner. The XDFile object

210

uses FileIO

212

to handle the file system transactions. By using the Database.Transaction file object, the XDFile object

210

handles database file transactions. The XDFile object

210

coordinates transactions for both the FileIO object

212

and the Database.Transaction file object

214

to keep both synchronized and to handle failure should it occur.

The UserData object

230

holds user data for a session of the X:Drive system. A session is basically a span of time for which a user engages the X:Drive system. Methods are included in the UserData object

230

to manipulate the user status, so that the activity may be monitored, as well as whether or not the user has logged in.

The SessionSecurity object

224

uses web logic session mechanisms to create the UserData object

230

. It does this by returning a separate object

226

. The SessionSecurity object

224

authenticates a user's login and expires old sessions with re-direction of such old sessions to appropriate pages.

The FileAction object

220

may have children classes and contain logic for determining request types such as user requests, administration requests, etc. Tests for file action requests such as quotas and permissions, etc., may also be handled by the FileAction object

220

. The FileAction object

220

accesses the file methods in the XDFile object

210

.

The XDError object

240

reads a configuration file of error lists which gives each error an I.D. number. Such error lists preferably pivot on the language in which the X:Drive system

100

of the present invention is programmed. Such lists should also be able to pivot on the partner with which the X:Drive system

100

operates. Default values for the lists may be to X:Drive errors in the English language. The XDError object

240

preferably holds errors in a stack and returns any such errors from the stack. Additionally, the XDError object

240

preferably accepts new errors by code or by message.

The XDXML object

242

accepts an object and delivers as output an XML representation of a transaction or status requested by the user or client software.

FIG. 3

shows the data flow through the X:Drive system

100

of the present invention, particularly that as reflected by the tiered configuration shown in FIG.

1

. From a starting point

300

, a request is sent by HTTP POST/GET command at step

302

. Web-DAV protocol may also be used and is currently considered preferable. The send request is implemented on the client

102

and is evaluated by the web server

120

as a request for static content in step

304

. If the request is for static content, the file is served by the web server

120

at step

306

, and the file is displayed at step

308

by the client

102

.

If at step

304

the request for static content is evaluated as negative, a proxy request is issued by the web server network

120

to the Java® application cluster

122

at step

312

. The request is received by the Java® application cluster (JAC)

122

and submitted to a servlet at step

314

. The Java® application cluster (JAC)

122

then parses the request header at step

316

. The Enterprise JavaBean™ (EJB) network

124

then authenticates the request at step

318

. If authentication cannot be achieved, process control is then re-directed to the re-login page via the JAC network

122

at step

320

. If authentication succeeds at step

318

, the JAC network

122

then parses the multi-part form data at step

324

.

The JAC network

122

then determines the type of request at step

326

. The request is then submitted to the FileAction EJB

220

at step

328

. The EJB network

124

then evaluates the request at step

330

in order to ensure that all the business rules and other applicable limitations are met, such as quota limitations, permissions, and the like. If the evaluation is successful at step

330

, the EJB network

124

then submits the request to the XDFile EJB

210

at step

332

and on to the transaction processor

146

. The appropriate actions are then taken via the transactional database

152

and the disk arrays

150

. If the business rule evaluation

330

fails, an error may be generated and, as for other errors in the data flow process of

FIG. 3

, a session error object

334

may be generated in a session error stack

336

.

In effecting the data transfer to the ultimate system resources

104

, evaluation is made as to the operation in step

340

. If the operation is not a data read operation such as a directory listing or file read, the error stack is checked at step

342

. If an error has occurred, the error status is sent to the client

102

at step

344

. The client

102

then accepts the transmitted XML code and renders the appropriate display for the user at step

346

. If the error stack evaluation step

342

does not reveal any error, a success message is generated at step

350

, and the subsequently-generated XML is received by the client

102

and displayed by the user at step

346

.

If at the evaluation step

340

, the operation is not a data read action, the error stack is checked at step

352

much in the same way as it was at step

342

. If an error has occurred, the error status is sent to the client

102

at step

354

. The error status message is then received as XML code by the client

102

at step

346

and displayed to the user. If at evaluation step

352

the error stack reveals no errors, the evaluation is then made by the EJB cluster as to whether or not the operation is a file read at step

360

. If the operation is a file read, the data stream is converted to a network stream and transmitted as a file to the client

102

by the Java® application network

122

at step

362

. The data is then accepted by the client

102

and served to the user at step

364

.

If at evaluation step

360

the operation is not a file read (see FIG.

4

), then by elimination, the action is a request for file metadata such as a directory listing indication of file attributes or the like. At step

366

, the metadata retrieved from the database

152

is then translated into XML format by the EJB cluster

124

. The XML data is then transmitted to the JAC network

122

, which encapsulates the XML from the network and sends it on to the client at step

368

. The JAC network

122

then sends the encapsulated XML to the client

102

for rendering and display at step

346

.

As indicated in the description above with regards to

FIG. 3

, users utilizing the client system

102

to connect to the X:Drive system

100

do so via the public Internet and then submit requests and receive replies effecting or indicating the user's requests. Requests for file manipulations, such as uploads, downloads, copies, moves and updates travel through each functional layer of the X:Drive system

100

.

The core of the EJB cluster, and as indicated in

FIG. 2

, the XDFile EJB provides core effectiveness in the present X:Drive system

100

. The XDFile EJB

210

is a multi-tiered component. The X:Drive system

100

stores file metadata (such as directory structure, file name, file attributes, etc.) in the database

152

for fast retrieval, sorting, searching, linking, and other capabilities beyond standard file systems. The actual file data is stored by the X:Drive system

100

in network-attached storage units or storage area networks such as those shown in

FIG. 1

, the NFS disk arrays

150

.

To access files that exist in this hybrid environment (bifurcated between file information and file data), X:Drive uses the XDFile object

210

to manipulate both files and file data in two-phase committal transactions.

FIG. 4

shows the details of these transactions.

In

FIG. 4

, the XDFile EJB system

400

allows entry at any one of the five darkened triangles. If the action is to be a copy, entry is made at the copy entry point

402

. If the action is a file read, entry is made at the file read point

404

. If the action is a file write, entry is made at the file write point

406

. If the action is a file delete, entry is made at the delete point

408

. If the action is a file move, entry into the XDFile EJB

210

is at the move entry point

410

.

Beginning first with a file copy action beginning at the copy point

402

, the evaluation of the operation occurs at step

420

, where determination is made whether or not the action is a read transaction. If the action is a read transaction, program flow proceeds onto the read action and entry point

404

. The corresponding database action

424

is then taken. As the action is a read transaction, the corresponding database record is read and evaluation is made as to whether or not the database action, in this case read action, has been successful at step

428

. If the read action is not successful, the changes are then rolled back, if any, at step

432

. An error is then returned at step

436

and the XDFile object awaits further instructions. If the evaluation at step

428

regarding the database action was successful, action can then be taken on the actual file itself on the OS File System

204

at step

440

. In the present case, the FileOS Action

440

is a read action, and the file may be read into a temporary buffer or other memory space. The FileOS Action is evaluated for success at step

444

. If the FileOS Action step

440

was unsuccessful, a fatal error is returned at step

448

, and the changes, if any, are rolled back at step

452

. If the evaluation at step

444

was successful, evaluation is made as to whether or not the action was a copy read at step

456

. If the action was a copy read, return is made to the copy entry point

402

at step

464

in order to perform the write portion of the copy function. If the evaluation at step

456

indicates that the action was not a copy read action, evaluation is made at step

468

to determine if the action was a move/copy action. If the action was a move/copy action, control is then directed towards the move entry point

410

via step

472

in order to delete the original file as the success of the move/copy transaction at evaluation step

444

indicates the success of the file write step of the FileOS Action step

440

. Program control is then turned over to the move/action entry point

410

so that the original file may be deleted at its original location via the delete entry point

408

.

If the move/copy evaluation step

468

indicates that not only was the action not a copy read, it was also not a move/copy, then the action is committed to the system at the ultimate system resource level

104

at step

480

and an indication of success is then returned at step

484

.

Upon reaching the move entry point at

410

, evaluation is made at step

490

to determine whether or not the transaction is a copy transaction. If it is a copy transaction, the program then enters and executes the copy entry point

402

. If not, the delete entry point

408

is activated to effect the remainder of the move transaction.

Consequently, it can be seen that a variety of actions take place depending upon the state of the XDFile EJB

210

at the database action

424

and FileOS action

440

steps.

In performing file reads and writes, simple one-step actions are taken because neither of these read or write actions are either copy reads

456

or move/copy

468

and so they fall into the system commit

480

and return a successful indication at step

484

. The same is generally true for the one-step delete action. Consequently, whenever a user wants to read, write or delete a file, entry can be made into the respective entry points at

404

,

406

, and

408

. Errors are returned when necessary.

However, the copy action

402

and the move action

410

require multiple loops through the XDFile EJB

210

in order to effect their operations. For the copy function

402

, the initial read must be made successfully with the evaluation step

456

then prompting the write step to occur by the return to the copy entry point at step

464

. The read transaction step

420

is then evaluated in the negative and the write entry point/action

406

is invoked with the database action occurring at step

424

to write the new information to the transactional database

152

and, if successful, the FileOS write action for the data at step

440

. If the file write is successful, the evaluation at step

456

as to whether or not the action is a copy read is answered in the negative as is the evaluation of the transaction as to whether or not is a copy transaction executed under the move action at step

468

. The resources are then committed, temporary resources are released, and the success indication is returned at step

484

.

Consequently, for a copy transaction

402

, the loop is first made through the read function

404

and then the write function

406

. For the move action at entry point

410

, a copy transaction is first executed with the two-loop operation as set forth previously. Upon completion of the copy action, the delete action

408

is implemented in order to erase the original file and its file data. Upon the third loop through the delete step

408

, the transaction is neither a read under the copy command at step

456

nor a copy under the move command at step

468

. Consequently, the move function has successfully completed, the system resources are committed at step

480

, and a success indicator is returned at step

484

.

In

FIG. 5

, an overview of the Java® architecture of the X:Drive system

100

of the present invention is shown. The Java® architecture

500

shown in

FIG. 5

may generally arise from the client

102

. A file action container

504

has certain attributes and operations as do the other beans of the architecture

500

. Contained within the file action container

504

are a number of stateful, stateless, and entity beans, as well as other containers having other beans. The file action container

504

contains two stateful beans: a user date stateful bean

506

and a process request stateful bean

508

. The user data stateful bean

506

has a user info entity bean to

510

and a security stateless bean

512

.

The process request stateful bean

508

contains a single container, the XDFile container

520

. The XDFile container

520

contains three (3) beans and a container. The three beans of the XDFile container

520

are: a database

10

stateful bean

522

, a file IO stateful bean

524

, and an admin stateful bean

526

. The container is a recovery container

530

which contains a recovery IO stateful bean

532

, a mount status stateful bean

534

, a recovery admin stateful bean

536

, and a recovery process stateful bean

538

.

As indicated by the nature of the beans carried by the containers, stateful beans generally carry information about the state of the bean, process, or otherwise as useful information for the ends and operations of the X:Drive system

100

of the present invention. Stateless beans generally carry no state information, and entity beans are generally for information or identification only. As Java® beans are objects intended to carry both data and processes in association with one another, it is up to the operations of the X:Drive system

100

of the present invention to selectively and appropriately activate the beans and enable the proper actions to take place. The file action container

504

is shown in alternative representation in FIG.

6

. In

FIG. 6

, a client

102

issues a user authentication request

602

and an operation request

604

. The user authentication request

602

is passed into the user data stateful bean

506

in the file action container

504

. The operation request

604

is passed into the process request stateful bean

508

. The user information entity bean

510

then transmits information to a user information database

610

, as does the security stateless bean

512

. The process request stateful bean uses a first property file

612

that is loaded upon deployment of the XDFile container

520

. The property file is loaded into the admin stateful bean

526

for use with the OS file system

204

. A Java® transaction server

620

may operate in conjunction with the database

152

as well as the OS file system

204

in order to process the operation request

604

. The second property file

630

may be loaded by the recovery admin stateful bean

536

upon the bean's deployment. The recovery IO stateful bean

532

and the recovery admin stateful bean

536

both transmit information to the recovery queue storage buffer

640

. The mount status bean

534

operates in conjunction with the mount status of the system

650

.

The recovery container

530

is called when once a failed resource begins to recover. Further description of the recovery process is given below. However,

FIGS. 5 and 6

operate in tandem to show linearly (

FIG. 5

) and organically (

FIG. 6

) the structure and operation of the XDFile object

210

.

FIG. 7

shows the detail of the XDFile database component. A transaction processor (such as Tuxedo from BEA) works in conjunction with the database transaction object

214

as well as the FileIO object

212

to provide a robust and reliable system. Both the database transaction

214

and the FileIO

212

objects include logic and/or programming to handle situations where database or disk array access cannot be guaranteed. The database.transaction object

214

handles the inherent doubt present in the system by using replicated or repeated clusters of databases. The replication process creates latency or delay, in the system. In order to accommodate this latency, the database transaction object

214

uses a session object (a data construct representing a user session on the X:Drive system

100

) to determine if the user's request can be transferred, or replicated, from one database cluster to another, in case of future system failure.

An important aspect with respect to the reliable operation of the X:Drive system

100

is the need to separate databases into functional groups. While the query database may be optimized for quick and small queries and while a transaction database might be optimized for fewer, larger, more time consuming updates, the database layer

236

in the X:Drive system

100

allows for associating SQL commands with different database clusters based on functionality. Additionally, the X:Drive database layer

236

is configured for consolidation and addition of databases on the fly.

As shown in

FIG. 7

, the SQL command

710

is issued and passed to a SQL command evaluator

712

. A SQL evaluator determines the SQL type so that the SQL can be sent to the appropriate database type (that is, in the X:Drive system

100

, the transaction database

150

, the query database

152

, or both).

Upon determining the database type of the SQL statement

712

, the database preference is evaluated at step

714

to determine if the user should be sent back to the same database. If the user is not to be sent back to the same database, the database currently bearing the least load is found in step

716

, and query is then made in step

718

to ensure that the selected least-loaded database is still up, running, and available. If it is, a specification regarding the pooling of database resources is created

720

and transmitted to the database object

236

. Database object

236

then takes the SQL command and passes it to the appropriate database, either the transaction database

150

or the query database

152

via associated connecting pools

730

.

If at step

718

the least loaded database is not available, an alternative database must be used and query is made at step

736

to determine whether or not the alternate database is up. If the alternate database is not up and the evaluation step

736

fails, additional databases may be queried or, as indicated in

FIG. 7

, a fatal error may be generated at step

738

. If the alternate database is up, a pool specification

720

is generated and passed to the database object so that the SQL command may be implemented upon the transactional

152

databases via the connection pools

730

.

If at step

714

the user must be sent back to the same database, query is made at step

740

to determine if that database is still up. If it is, the request is passed to the pool specification

720

where it is subsequently passed to the database object

236

, on to the connection pool

730

, and the appropriate database, either the transaction database

150

or the query database

152

. If the same database is not up and the evaluation at step

740

fails, an alternative database must be used, but the SQL request is queried at step

744

to determine if the SQL command is transferable to the alternate database. If not, a fatal error occurs at step

746

. If the SQL command is transferable, query is made at step

750

to see if the alternate database is up and active. Should the evaluation fail, subsequent databases may also be queried if the SQL command is transferable. However, as shown in

FIG. 7

, if the second database is unavailable, a fatal error may be generated at

746

. Otherwise, the database is up, and the evaluation at step at

750

is successful and the command is made available to the database object

236

via the pool specification standard

720

and on to the databases through the connection pools

730

.

In order to ensure proper operation of the XDFile database object

210

, a database status monitor

760

persistently and on-goingly queries the databases

150

,

152

. The status is then returned to a database status object

762

. the database status object may provide information to the recovery container

530

of the XDFile object

210

.

The recovery mechanism for the X:Drive system

100

of the present invention is shown in FIG.

8

. The FileIO object

212

uses a recovery object such as the recovery container

530

to handle write transactions

406

(as opposed to read transactions

404

) when the transaction processor

214

fails. The recovery object is transparent to the user, making it easier and more convenient for the user to use the X:Drive system

100

while decreasing the concern that such a user would have in case of a power outage or other failure in one part of the X:Drive system

100

.

The FileIO object

212

reports an error to the user, but informs the user that her request was stored in the X:Drive system

100

and that the X:Drive system

100

will try to apply the change as soon as possible. If the storage unit, represented as a mounting point in the EJB cluster becomes unavailable for write transactions

406

, the monitoring client

760

updates the EJB network

124

that the status of the mounting point is “down.” Once the mounting point is available and checked for data integrity, the status is updated from “down” to “recovery” and the recovery object

530

is called to apply all queued requests for the file action container

504

. This keeps the user from catastrophically losing uploads and other file writes, but may cause some delay in file reads.

In the recovery system

800

of the present invention, the multi-connected pooled database object, the recovery-enabled FileIO object

212

, and the transaction processor

146

work together to create a resource layer offering high availability, recovery, and scalability. Additionally, this resource layer (encapsulated in the XDFile EJB

210

) lends itself to replication of the data, both geographically and locally. Such replication preferably has the three essential traits of being off-site, application-driven, and accessible. With this level of controlled replication, secondary X:Drive clusters are enabled in geographically diverse locations in order to enhance the reliability of the X:Drive system

100

. Consequently, data loss from one data center or even the physical loss of an entire data center would not cause loss of customer data or access. Re-direction would occur dynamically and this information would be replicated in a plurality of sites across the X:Drive system

100

, the query or metadata databases provide multiple pointers to the user's data.

In the recovery system

800

of

FIG. 8

, the recovery system is initially initiated when the MPS Bean

534

is set for a mode to detect mount point recovery at step

804

. At step