PrizmDoc Viewer v13.25 Release - Updated
PrizmDoc Viewer / Administrator Guide / PrizmDoc Application Services / Clustering / Optimize Cache Performance for Cluster Environments
In This Topic
    Optimize Cache Performance for Cluster Environments
    In This Topic

    Introduction

    PrizmDoc Application Services (PAS) utilizes cache locations that don't need to be on the same server. If multiple PAS servers are in use, they can be configured to use the same central filesystem and database so that cached data is shared between servers.

    A given Viewing Session is cached by either PrizmDoc Server or by PAS depending on how it was created. Sessions created using a documentId are stored in PAS' central cache as a pre-converted viewing package. As these sessions are not cached in PrizmDoc Server, they will not be directly accessible through the PrizmDoc Server API.

    Example

    POST http://localhost:3000/ViewingSession Content-Type: application/json
    {
        "source": {
            "type": "document",
            "fileName": "sample.doc"
            "documentId": "doc_9495837910qc"
        }
    }
    
    

    Viewing packages can be created either explicitly via the Pre-Conversion API or, as in the above example, implicitly by providing a documentId in the Viewing Session API.

    Sessions created without a documentId are not stored as viewing packages and are cached by PrizmDoc Server as normal according to the server configuration and request parameters.

    Example

    POST http://localhost:3000/ViewingSession Content-Type: application/json
    {
        "source": {
            "type": "document",
            "fileName": "sample.doc"
        }
    }
    
    

    NOTE: If PrizmDoc Server is running in Cluster Mode, PAS handles the use of affinity hints internally and does not require the user to perform anything additional for optimized PrizmDoc Server cache performance.

    You can find more information about PrizmDoc Server, PAS, & Cache in the Implement Caching Strategies topic.