REAP Cataloging and Searching

From OPeNDAP Documentation
Jump to: navigation, search

1 Summary

To extend the existing metacat-based search system that Kepler uses so that it can be used to find DAP servers, we will:

  1. Use the AIS capabilities and Aggregation to be developed for Hyrax 1.6 and/or already present in TDS to add specific information to the Ocean Use Case data sets.
  2. Build a module that can transform the DDX returned by these servers into an EML document
  3. Store those EML documents in an instance of Metacat that Kelper can serach
  4. Modify Kepler to search over EML that contains a <spatialRaster> element.

1.1 Related designs/projects

See also a page that documents the Current state of the THREDDS Crawler on which this depends.

2 Use Cases

2.1 Setup the catalog infrastructure

2.2 Add information about a data set to the catalog

2.2.1 Background information

Sample EML for CZCS data

Parts of an EML document that must be built:

  • Title
  • Optional: Abstract
  • Creator
  • KeywordSet
  • Contact
  • For Ocean Use-Case data sets: SpatialRaster
2.2.1.1 Regarding the origin of the SpatialRaster element in EML

The SpatialRaster element in EML has it's origin in the FGDC CSDGM (Content Standard for Digital Geospatial Metadata). However, it has a close relative in the ISO 19115 standard and XSLT can be used to transform between the EML, FGDC/CSDGM and ISO 19115 elements (or 'elements' in the case of a Content Standard).

The vector metadata information in EML is based on ISO 19115.

2.3 Search the catalog

Needs significant work


3 Definitions

4 Background

For systems like the ones which use DAP the biggest data location problems are getting valid metdata that is sufficiently uniform and making a smooth transition from the initial process of location to the selection of individual parts of a chosen data set.

4.1 The problem of heterogeneous configuration

The first problem is the really the problem of finding data sets in the vernacular sense of finding and data set while the second is the problem of taking a number of ad hoc heterogeneous storage and organizational configurations and mapping them into a (mostly) uniform access mechanism. While both problems present a significant challenge, the second can be effectively addressed by aggregating the discrete elements of the actual data set's configuration (e.g., 20,000 images in a file system where different years and months are each in nested directories) into a single logical entity so that the different discrete parts are accessed in one operation. An example will make this clearer: Imagine the collection of 20,000 images stores data in directories named 1999, 2000, ... and within each of those there are sub-directories named 01, 02, ..., 12 and within those there are files named following the pattern YYYYMMDDD where DDD is the day number. Lets also assume that this data set is served by Hyrax and thus each of these files can be accessed by a single URL. If a person wants to read data from the fourteen files from Jan 25th to Feb 7th, they have to know this structure and apply it to the data set to figure out which URLs to use with their client, then get the data from the URLs and probably assemble it into a three dimensional data structure. If all collections were organized like this, clients could be built to perform those operations and the problem would be solved. But the variability among data set storage patterns is actually very high - a handful of data sets are stored as described by this example, but most are stored in other ways and there are enough of those 'other ways', and new 'other ways' keep emerging every day, to make customizing clients impracticable.

A solution is to aggregate the different URLs into a single three-dimensional data object and provide a data server than can operate on it without revealing its true composition. In this scenario, the person who wants data from the Jan 25th to Feb 7th asks for it by accessing the data set using the uniform data access operations supported by every data server within the system. For 100 different data sets made up of images taken over time, each can represented as a single three-dimensional data set and each can be accessed using the same operators (and thus operations) even though they are actually stored using 100 different configurations of files and databases. This is how the process of aggregation can be used to solve the data location problems that arise from heterogeneous configurations of data sets.

The downside to this approach is that it applies only to collections that can be aggregated. If a dataset is composed of many files both those cannot be aggregated using one of the schemes provided by NcML, then this technique won't solve the problems associated with making the transition from finding a dataset to selecting elements of its inventory.

The THREDD Data Server (TDS) supports both DAP and aggregation using NcML and clearly shows that this approach works well in a wide variety of cases

4.2 The problem of finding data sets

This is the problem of matching a group of data sets to a schema that uniformly describes their variability. The entries that conform to this schema (aka records) cover both taxonomies and values for sets of predefined parameters. The problem here is one of heterogeneity too. Clients are optimized to search for information specific to a fixed set of problem areas (e.g., Ocean data from satellites; Biological Ocean data from fixed locations) and because of this, are built to work with specific search information. Each data set is described by a record and the collection of records is stored in a data base that can be searched. There's nothing particularly hard about this and it works well with one caveat: The form and context of the records tends to be very specific to a problem domain. Of course, information about multiple domains can be included in the same record; it's just as easy for a data base to search the expanded records as for the focused ones. However, the problem lies in the making of those records. If the information they contain is very tightly focused so that only one problem domain is described, they are easiest to write but useless when searching within other domains. If the information is universal the records are essentially impossible to write. The middle ground is always a compromise between breadth of coverage and manageability. In systems built using DAP servers, there's never a requirement to add specific metadata to start serving data so the caliber of information useful for searching is highly variable. We chose this limitation because it was the best way to get the most data served. (And because there are many cases where network access to data is desirable within a group where many metadata parameters are well known).

The problem of metadata development needs to be separated from the production of the data themselves because as the data age their potential audience tends to widen. At first only those most closely connected to the data are interested in it, but over time interest often broadens until people who initially know very little about the data become potential users. As this widening of scope takes place the need for more general metadata increases. So too does the variety of clients which might be used to operate on the data. In other words, as the pool of users widens, so does the kinds of uses and as that happens the problem domains where the data may be used widens.

The challenge for a good searching system is to accommodate these changes over time. The system must support building additional metadata into the network presentation of data over time. To do this effectively, the system needs to build different types of records using more generic 'building blocks' which can be shared between several different types of records used by different types of clients.

The AIS- and Aggregation-based system described here is designed to solve both of these problems.

5 Design

Components of the Proposed REAP Ocean Use-Case Catalog System

The Kepler workflow client used in the REAP project has the ability to access data using DAP servers but has no way to find those servers. Data search systems built for DAP servers don't have a very good track record because such systems (for the most part) do not address the twin needs of working with a fluid pool of data servers and data sets and fitting in with the basic requirement of DAP-based systems - that impact on a data provider be absolutely minimal.

In order for the impact of hosting a DAP server to be minimal, the typical data documentation (i.e., metadata) required for most searching systems is not required for data served using DAP. As a result, interfacing DAP servers to such systems is a daunting task involving lots of manual metadata entry. This effort is frustrated not only by the often baroque nature of metadata standards (e.g., FGDC) but also because the data sources being described move from place to place frequently, something the begs for automated discovery and cataloging - exactly the opposite of what is provided by hand-written metadata records.

Complicating this typical scenario is the nature of most searching systems: They tend to be tailored to a specific client system. Those data providers remaining who were not deterred by the issues of complexity and frequent manual updates of metadata often are when they see that the additional metadata will satisfy the needs of only one client. It is virtually impossible to get data providers to write these metadata records, since the providers will have to write differently-formatted information for each such client. This would normally be remedied by adopting a standard and then modifying all clients to use that standard. However, standards with enough breadth to satisfy the semantic needs of a wide spectrum of clients are, as previously described, very complex.

Since it's unlikely that a 'magic' standard will appear anytime soon or that clients will drop many of their requirements for metadata vis-a-vis searching, we need a solution that will provide a way to build a uniform set of metadata at the servers which can then be assembled by different clients according to their diverse needs. It might be that some desired information is missing from some servers or some superfluous information is present at others, but the clients can build their choice of metadata records using what is present.

The companion technologies of XML and XSLT combined with a system to provide ancillary information for data sets served using DAP suggest one solution to this problem.

The system described here will use Kepler as an example client. It will build metadata records using EML, where most of the really important metadata is actually a series of XML micro documents made up of information described by ISO 19115:2003, 19115-2:2009 and other standards. The system will use a server-side solution (technology leveraged from other projects and thus more likely to be in use) to augment data sets with this information. The EML documents will be built using XSLT - EML won't be directly returned by the DAP servers and the same geo-spatial information can be used for other things (e.g., WCS 1.x). The EML will be scavenged by Metacat, which Kepler already knows how to use (with some caveats). Metacat doesn't know how to crawl DAP servers, but it can be fed URLs and we may employ TPAC's crawler to feed Metacat with URLs or DDX/EML objects.

This solution is not ideal. Data providers will still initially have to write metadata records, albeit smaller, more concise ones. However, automated crawling of servers and automated harvesting of the discovered URLs means that data set and server movement can be accommodated more effectively than with designs based on static documents.

A problem with this design is that data providers will need to know the collection of 'micro documents' needed to support one or more different client systems. We could mitigate this risk by surveying clients to find out how diverse their needs really are - not in terms of formats but in terms of content. We know from current experience with XSLT and related technologies that we can transform information stored in XML fairly easily, so supporting different textual formats is not nearly as much of a concern as differing content requirements.

5.1 Kelper Search Interface

Mock up of a search interface, originally made for the DAP Matlab Toolboxes built at URI. We can use this as rough schematic for the searching needs of the Ocean Use Case within REAP

In order to build a complete end-to-end system to search for DAP data within Kepler, and the Ocean Use Case in particular, Kepler must be modified to search over EML's <SpatialRaster> element (or the equivalent) and it must have an interface that will provide a way for users to make those searches. The mock up to the right is one way to do that.

The interface shows that the criteria needed for the Ocean Use Case are:

  1. Geographical location, as a box in latitude and longitude
  2. Time range
  3. Resolution
  4. Parameter, where the list of parameters is:
  • SST, SSH, Chlorophyll, precipitation, Color, Vector Wind, Vector Stress

5.2 Data server URLS for the REAP Ocean Use Case

5.2.1 GHRSST data at NODC

There are a total of 16 data sets (I think). Many of these are updated daily or at least often.

Top level; there are a number of products under this.
http://data.nodc.noaa.gov/opendap/ghrsst/
  • Within .../ghrsst:
  • L2P/GOES11
  • L2P/GOES12
  • L2P/SEVIRI_SST
  • L2P_GRIDDED
  • L4

Results: 150,000, 31 classes using the three-rule classifier.

5.2.2 URI 1-km data sets

Results: 159,206 DDX URLs; 15h 26m; 41 classes using the three-rule classifier.

5.2.3 Goddard SFC Servers

5.2.4 PMEL

Hankin's site: http://ferret.pmel.noaa.gov/thredds/dodsC/data/PMEL/ Not sure how you really crawl this site, but since I don't exactly what your crawler does maybe it is easy or you can figger it out.

5.2.5 PFEL

Mendelssohn's site: http://oceanwatch.pfeg.noaa.gov/thredds/catalog.html

5.2.6 Hawaii

site: http://apdrc.soest.hawaii.edu:80/dods/public_data/. Again, this one is very well organized, but may be in a form that precludes crawling although I'm pretty sure that they are using a TDS.

5.2.7 IFREMER

I don't know how to access this site. Here's a URL for some data in the site: http://www.ifremer.fr/thredds3/standard/dodsC/ODYSSEA-SST-NRT_VIEW_BEST_ESTIMATE.html, but I can't go up a level. Not sure what's up here. And then there's the MyOcean site. I can send a message off to Jean-François Pioli if you like?

5.2.8 HyCOM

For now we will pass on the HyCOM data sets because the server seems broken.

5.2.9 CSIRO

I'm not sure if the use-case uses data from CSIRO, but here's one URL to data they have: http://opendap.csiro.au/opendap/

6 Risks

  1. In order for this to work, Kepler's search over the EML records has to be modified so that it includes using the information in a <spatialRaster> element.
  2. This design does not address how Kepler users would actually perform a search (it does not specify a User Interface). Since the existing Kepler search over EML records appears to return a set of results where each data set is a different actor, it's not clear how to make this mesh with the DAP actor, where a specific data set (i.e., URL) is a parameter to the Actor. Maybe have the search return a set of Actor 'objects' each of which has the URL parameter bound to a value?
  3. Even though the design tries to minimize the amount of new metadata to be written, the reduction might not be enough to be a viable solution.
  4. One aspect of the problem identified in the design overview is that data sources tend to 'move' over time and any general solution will need the ability crawl host/domain names and find moved data sources. This design lacks a crawling component. That should not be an issue for the Ocean Use-Case data sources, but it limits the solution in the general case. A potential solution exists in the TPAC crawler.

7 Deliverables

  1. The Guide: A manual for the DP that provides information about what fragments to add to a data set to support a specific metadata record type. Initially this will support EML records that can be used in Kepler
  2. A modification of Kepler so that it will search for geo-spatial data.
  3. The NcML-AIS (implemented as a module for the BES). Done 10/1/2009
  4. The NcML-Aggregator (implemented as a module for the BES, likely combined with the AIS) Partially complete 1/1/2010. We can perform aggregations of Grids using the JoinNew strategy and NcML's dataset scan feature is mostly implemented but lacks the 'date' feature. Likely completion is 2/1/2010.
  5. A modification to the DAP (implemented in libdap++) that provides a way to insert XML into the DAP variable attributes Done 3/2/2009
  6. A crawler to harvest DDXs that will be transformed into EML. Partly completed 3/5/10. The code is described above under THREDDS Crawler
  7. A tool that can produce a metadata record of type X for a given data set (DAP URL). Likely based on XSLT
  8. A tool, bundled with the Guide, to test the acceptability of the metadata generated by the system (i.e., is the resulting EML document not only valid EML but will it work in the Kepler client).

8 Period of use

This will be initially developed for use by the REAP project, which has a period of performance that ends in May 2010.

Some elements will be part of a production version of the BES (the NcML-AIS/Aggregator) while other parts are primarily proof-of-concept and will not be useful beyond the REAP project (the Guide sections regarding EML). Hopefully the XSLT will survive and prove useful in other applications.

9 Appendix

9.1 Setting up MetaCat on OS/X

Using OS/X 10.5.8

9.1.1 Prerequisites

9.1.1.1 Java 5.0

This is present on OS/X 10.5.8. Set JAVA_HOME to /System/Library/Frameworks/JavaVM.framework/Versions/1.5/Home. If java 1.5 is the default, then the shorter version /System/Library/Frameworks/JavaVM.framework/Home will also work. Set JAVA_HOME in both .bashrc and environment.plist

Here's a reference on stuff about JAVA_HOME: http://www.oreillynet.com/mac/blog/2001/04/mac_os_x_java_wheres_my_java_h.html and Apple's technical documentation on Frameworks: http://developer.apple.com/mac/library/documentation/MacOSX/Conceptual/BPFrameworks/Frameworks.html

9.1.1.2 Postgres

Get Postgres from http://www.postgresql.org/download/macosx and install. The only trick here is that the Mac is setup with too little shared memory for postgres to work well, so you need to create a sysctl.conf in /etc/ and restart the machine. Here's what I put in mine:

kern.sysv.shmmax=1610612736
kern.sysv.shmall=393216
kern.sysv.shmmin=1
kern.sysv.shmmni=32
kern.sysv.shmseg=8
kern.maxprocperuid=512
kern.maxproc=2048
9.1.1.3 Ant

This is needed only if you will be building Java from source. But it is really easy, so ...

Get Ant from Apache, install it somewhere sensible like /usr/local and set ANT_HOME to point there in both .bashrc and environment.plist.

9.1.1.4 Tomcat

Same as Ant above, but set the environment variable CATALINA_HOME.

I installed both 5.5 and 6.0, just to see if metacat will work with 6.0.

9.1.1.5 Apache

OS/X comes with this. Here's a set of tutorials from the most basic starting point: http://onlamp.com/pub/ct/49. Short version: Apple menu --> Preferences --> Sharing --> Web Sharing (turn it on).

The command httpd -V returns a fair amount of information; The root directory of the server is /etc/apache2; the modules are in /usr/libexec/apache2; and the document root for the machine is /Library/Webserver/Documents/.

First get metacat running with postgres & tomcat and then look at apache. Likely we can use the proxy and rewrite modules in place of mod_jk, thus avoiding the mod_jk hassles like compilation.