Lessons learnt on distributed viewing services Lounaispaikka Map Service and ICZM Map Viewer of the ENVIFACILITATE

1
Lessons learnt on distributed viewing services
Lounaispaikka Map Service and ICZM Map Viewer of
the ENVIFACILITATE

• Antti Vasanen
• Regional Council of Southwest Finland

2
Lessons learnt on distributed viewing services

• In this presentation I will concentrate on the
  characteristics of distributed data management in
  internet map services
• I will use the Lounaispaikka Map Service and the
  ICZM Map Viewer as examples, both of which have
  been developed as a part of the ENVIFACILITATE
  project
• Overview of the presentation
• Introduction to the map services used as examples
• Description of the service architectures of these
  services
• Lesson learnt on utilising distributed viewing
  services

3
Lounaispaikka Map Service

• The Lounaispaikka Map Service is a central
  component in the GI services of Lounaispaikka,
  the regional GI service and network operating in
  Southwest Finland
• In the recently renewed map service, all the
  previously separate map services of Lounaispaikka
  have been merged together
• The trilingual viewing service includes a wide
  variety of different thematic maps
• The easy-to-use service interface takes into
  account the needs of both GI professionals and
  everyday users with no prior GI knowledge

4
(No Transcript)
5
ICZM Map Viewer

• ICZM (Integrated Coastal Zone Management) is an
  European initiative to integrate different
  policies which have an effect on the EUs coastal
  regions
• The ICZM Map Viewer aims to provide an
  easy-to-use interface to relevant spatial data
  sets required
• to understand the environmental conditions and
  land use pressures on coastal regions at an
  international scale
• to understand and join together the (conflicting)
  needs of coastal land use and management
• to support the elaboration of coastal strategies
  in Finland, Estonia and Latvia
• The Map Viewer is available in English and it
  covers the coastal areas of Finland, Estonia and
  Latvia

6
(No Transcript)
7
Distributed Map Services

• Traditionally internet map services have included
  all the needed data on the same server as the map
  service operates
• Unlike centralised services, distributed map
  services contain a portion of the data on two or
  more servers
• Both the Lounaispaikka Map Service and the ICZM
  Map Viewer utilise distributed methods in their
  data management

8
Service Architecture of the Lounaispaikka Map
Service

• Most of the data situates on the map server of
  Lounaispaikka
• This is due to the lack of operators serving data
  in a distributed manner in Finland
• Thus, the centralised service architecture is
  currently dominant in most internet map services

9
Service Architecture of the Lounaispaikka Map
Service

• The standardised WMS interface enables the map
  application to import GI data directly from the
  data provider
• The Lounaispaikka Map Service includes WMS layers
  from e.g. NLS, FMI and GTK
• In Finland, WMS has been adopted as the main
  mechanism for creating the national map service,
  and thus, most of the data will come from
  national WMS services in the future

10
Service Architecture of the Lounaispaikka Map
Service

• Distributed database connections are currently
  used only to obtain registered bird observation
  data from the database of the BirdLife Finland
• Database connections are also used in the Marine
  Environment service, but the databases are
  currently located in the database server of the
  University of Turku

11
Service Architecture of the Lounaispaikka Map
Service

• Metadata is currently obtained to the
  Lounaispaikka Map Service from the Lounaispaikka
  metadata catalogue
• In the future, when the national infrastructure
  becomes more developed, metadata information can
  hopefully be acquired directly from a common
  national metadata catalogue

12
Service Architecture of the ICZM Map Viewer

• The ICZM Map Viewer uses the distributed data
  management methods offered by the ESRI ArcIMS
• Data covering all three countries are located on
  the ArcMap image service running on the
  ENVIFACILITATE map server

13
Service Architecture of the ICZM Map Viewer

• National datasets are situated on separate
  servers maintained by a ENVIFACILITATE project
  partner in each country
• Due to the lack of software resources, the
  Latvian ArcIMS service is temporary situated on
  the same server as the Finnish service.

14
Lessons learnt from distributed data management

• Data availability increases notably when level of
  distribution rises
• In addition to basic datasets, distributed data
  management enables the adoption of huge domestic
  and international data sources
• Lately, especially the number of WMS data sources
  has increased rapidly
• Lounaispaikka Map Service, for example, obtains
  species information from a Canadian WMS
  interface, which combines hundreds of data
  sources world wide from Global Biodiversity
  Information Facility (GBIF)

15
Lessons learnt from distributed data management

• Maintenance efforts decrease when level of
  distribution rises
• Maintaining map services and updating its data
  content requires most efforts after the actual
  development phase
• When using distributed data management, related
  efforts may decrease considerably
• However, if distributed data sources are numerous
  and potentially unstable, the overall usability
  of the map service may decline in spite of the
  decreased maintenance needs

16
Lessons learnt from distributed data management

• The need of server capacity is inversely
  proportional to the level of distribution
• In many cases, map services based on vector GI
  data need substantial server capacity to produce
  raster format maps readable on the internet
• Using distributed data management may accelerate
  the map service considerably

17
Lessons learnt from distributed data management

• Extensive possibilities for GIS functionality
  often requires that the data is situated on an
  internal map server
• If large scale GIS functionality is needed,
  simple image based distributed data management
  systems, such as WMS, will not be suitable
• However, vector based data transfer methods, such
  as WFS (Web Feature Service) may solve this
  problem as they become more widely available

18
Lessons learnt from distributed data management

• Control over the cartographic visualisation of
  the map content may decline notably when
  distributed methods are used
• The provider of the interface usually defines the
  visual expression of the distributed data
• Affects most image based systems such as WMS
• However, when using coordinate point based data
  transfer methods, such as distributed databases,
  the interface provider has basically no effect
  on the cartographic visualisation of the data

19
Lessons learnt from distributed data management

• Map service functionality may become more or less
  unstable if many distributed data sources are
  used simultaneously
• Depends on the used data management system
• In WMS services, a malfunction in the data
  providers server causes mere absence of the
  required data
• However, e.g. in distributed ArcIMS services, the
  whole map services will collapse in the case of
  the malfunctioning in one of its components

20
Conclusions

• Distributed data management is a real opportunity
  when creating a map service depending mainly on
  other data provides datasets
• Using standardised data transfer methods, such as
  WMS, is recommended for many reasons
• easy to initialise and use
• large quantity of data sources
• well supported in both commercial and open source
  applications
• In our experience, more stable than licensed
  products
• However, distributed data management may not be
  desirable if for example
• advanced GIS tools are needed
• the service provider has precise requirements
  regarding visualisation
• the map service is entirely based on the service
  providers own datasets

21
Thank you!
antti.vasanen_at_varsinais-suomi.fi lounaispaikka
.fi/map envifacilitate.utu.fi/iczm