gms | German Medical Science

GMDS 2013: 58. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS)

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie

01. - 05.09.2013, Lübeck

Hospital Information Systems on High Performance and Energy Efficient Database Systems

Meeting Abstract

Suche in Medline nach

  • Christian Thies - Reutlingen University, Reutlingen, DE
  • Ilia Petrov - Reutlingen University, Reutlingen, DE

GMDS 2013. 58. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e.V. (GMDS). Lübeck, 01.-05.09.2013. Düsseldorf: German Medical Science GMS Publishing House; 2013. DocAbstr.35

doi: 10.3205/13gmds130, urn:nbn:de:0183-13gmds1303

Veröffentlicht: 27. August 2013

© 2013 Thies et al.
Dieser Artikel ist ein Open Access-Artikel und steht unter den Creative Commons Lizenzbedingungen (http://creativecommons.org/licenses/by-nc-nd/3.0/deed.de). Er darf vervielfältigt, verbreitet und öffentlich zugänglich gemacht werden, vorausgesetzt dass Autor und Quelle genannt werden.


Gliederung

Text

Current Hospital Information Systems (HIS) face operational challenges such as long downtimes/data inconsistency during system updates, high energy costs, system slow-down during periods of high load and incomplete failover solutions with lots of costly and error prone manual post-processing. A bottleneck for all of these problems is accessing the HIS database and its consistency. To overcome these problems the software architecture (SWA) of HIS currently only uses DBMS specific features. Considering the most frequent types of transactions from a HIS the database itself typically has a monolithic design. Here a higher integration of hardware, DBMS and HIS architecture will increase efficiency, reliability and decrease HIS downtime. This is achievable by using recent development of new memory technologies and affordable hardware as well as a systematic overview of sophisticated database concepts and their specific integration in HIS architectures. This work describes the impact of such new concepts on the main challenges in HIS data management and HIS SWA.

Problem Definition: An in-depth analysis of existing production HIS identified the following challenges:

  • Flexibility - HIS frequently undergo schema changes in presence of large data volume. Ability to handle schema evolution gracefully, minimising the maintenance downtime is a critical requirement.
  • High update rate -- HIS produce bursty, write-intensive workloads resulting from the schema reorganisations and regular modifications.
  • Performance Monitoring -- Built-in nonintrusive performance monitoring is critical for timely performance bottleneck detection and proper HIS maintenance.

Two technical trends are looming: Energy-Efficiency (EE) and High Performance Computing. EE implies energy proportionality [5], [1] and efficient use of new hardware technologies [3].

Material and Methods: We propose alternative architectures based on alleviatating downtime costs by letting part of the HIS serve read processing, and delaying update processing while schema changes are applied on a HIS database snapshot. In an EE architecture we distinguish a fast update data store holding the up-to-date HIS state and a set of read only data stores. The HIS storage manager employs synchronisation mechanisms to ship the newest state to the read nodes and apply read optimizations. The read nodes are low-power micro or mini-servers (4W to 20W) built on low-power hardware for scale out [2], [4]. Once synchronised such servers can be turned on to handle the outstanding read transactions depending on the current workload. An alternative approach targets high performance systems with main memory DBMS and fast interconnect. The core idea is to handle expensive schema updates based on snapshotting to reduce downtime. Once a schema update is requested the second/update system is activated and fed with a snapshot form the operational system. While update is applied on it the operational system serves read-only transactions. Using this architecture the downtime is significantly reduced and plannable, while the system remains operational for read transactions.

Results and Discussion: Rethinking HIS SWA to face challenges like health record analysis requires modern database solutions and state-of-the-art hardware. Even during large system updates and migrations the HIS remains available for read access. Extended downtimes due to erroneous updates are avoided.


References

1.
Härder T, Hudlet V, Ou Y Schall D. Energy Efficiency is not Enough, Energy Proportionality is Needed! In Proc. DSFAA. 2011: 226–239.
2.
Intel Corp. Intel Atom Processor S1200 for Microserver: Datasheet. 2011; Vol. 1.
3.
Petrov I, Bausch D, Gottstein R, Buchmann A. Data-Intensive Systems on Evolving Memory Hierarchies. GI-Jahrestagung. 2012: 427–433.
4.
Raspbery. Raspbery Pi Datasheet. 2011. http://www.raspberrypi.org Externer Link
5.
Schall S Hudlet V. WattDB: an energy-proportional cluster of wimpy nodes. Proc. ACM SIGMOD ’11, New York, NY, USA. p. 1229–1232.