Caché Release Notes and Upgrade Checklist Archive
New and Enhanced Features for Caché 2012.2
The following major, new features have been added to Caché for this release:
Furthermore, this version of Caché has been improved and enhanced in the following areas:
In addition, many more localized improvements and corrections are also included. In particular, if you are upgrading an existing installation, please review the detailed list of changes in the Upgrade Checklist
Using Unstructured Data In Analytics
DeepSee support for unstructured data
DeepSee now supports unstructured data. DeepSee cubes can use free-text as the source for dimensions and measures. Using the iKnow technology, this text is analyzed, and the results of this analysis ( the "concepts" from the text) are available as members of the dimensions.
Cubes can contain a mixture of dimensions based on structured and unstructured data. Like all DeepSee cubes, these cubes can be queried via DeepSee Analyzer, dashboards, and programmatically via MDX queries.
DeepSee includes new Term List functionality. A Term List is a set of key/value pairs that can be used for lookups and translations. Term Lists can be used during cube building, and also in MDX via the %TERMLIST function. For defining Term Lists, there is a Term List Manager option on the DeepSee menu in the Management Portal.
Ensemble business rules can be used at cube build-time. A sourceExpression can use the new %Rule method to invoke a business rule. The business rule is passed an instance of the current source object, and the result of the business rule is returned.
This release introduces a new capability that enables Caché, Ensemble, and HealthShare users to interactively define reports via a drag-and-drop browser interface. This new technology, DeepSee Visual Reporting, consists of three parts:
DeepSee provides a browser-based UI for creating report data definitions. This is analogous to creating a query that gathers data for the report and specifying how that data will be formatted when it is displayed. The user creating the definition needs an understanding of the underlying source data and its interrelationships.
Once the data definition is complete, you can display the data in any number of formats. You define these layouts also via a browser-based interface that provides the ability to preview the output using the actual data chosen.
Once the user completes that definition of the data and output formats, the report can be run at any time by the author or those with whom the report is shared.
For 2012.2, the Report Data Definition and Report Format Definition components of DeepSee Visual Reporting are available for evaluation, but are not supported for production. These components will be fully supported in a future release.
Reports created with DeepSee Visual Reporting are supported in production in 2012.2.
Rapid Application Development
In this release, the iKnow API has been extended with the new Semantics package. This package contains two major new feature sets.
Semantic dominance is a term used to describe how important a specific element (can be a word, a concept, a set of concepts or a sentence) is in the context of specific text and in relation to the other information in the text. The semantic dominance API allows applications to identify:
the semantically most dominant elements in a single source, that is, the most dominant elements in a domain or a domain subset;
how dominant elements are shared by different texts; the most typical sources in a domain-the most atypical sources in a domain.
The uniqueness and added value of the semantic dominance API resides mainly in two elements. First, of all the algorithm implemented works based on the context of a single source only and doesn't need a reference corpus to identify the base line importance of a text element. Second, the algorithm generates values that are comparable between different texts and between different text elements. For example the semantic dominance value of a single concept can be compared with the semantic dominance value of a pathway consisting of multiple concepts and relations.
The semantic dominance API can be used to easily find out what the important common content of a series of text is or to find out what texts are the most typical for a given set of texts. It allows also to quickly find out what the important new information is in a text.
With the new capabilities in the proximity API, it is easy to find out which concepts appear in the context of a chosen reference concept. The functions in this API are used to find out what concepts are associated with the reference concept and how strong the association is. Based on the contents of the domain for which the semantic proximity is calculated the semantic proximity profile for the concept bakery will contain concepts such as muffins
, belgian waffles
each with a different value to express the strength of the association with bakery.
Beginning in this release, the default doctype for all Zen pages is HTML 5 for browsers that support HTML 5 and CSS 3. Also included is a new generated css file that will be served only when Caché detects an HTML 5 supported browser and the output contains CSS 3 syntax. You may override this behavior by changing the %OnDetermineCSSLevel()
callback to return 2 instead of 3 or by modifying a global node in the managers database. With this change, we also start using the native SVG renderer in IE 9 and above.
Improved Arabic Rendering In Reports
Starting with this version, Caché officially supports Arabic output in Zen Reports using Apache FOP.
With 2012.2, InterSystems introduces a feature-rich alternative to ObjectScript pattern matching, Regular Expressions. Using Regular Expressions not only detects if a complex pattern is present, but also provides location and results of successful matches. This allows more complex manipulation of strings in less code. For details on usage, please refer to the class documentation for %Regex.Matcher
As part of version 2012.2, InterSystems offers a simple way for application developers to dynamically change the memory allocation for processes during runtime, or define large memory allocations during process startup time. Available system memory can be by fully utilized to perform complex and high performance operations by keeping arrays in memory.
While this feature is most useful on 64bit operating systems, process private memory can also be changed in 32bit environments. Customers need to verify that use of this feature does not trigger unexpected performance profiles in the operating system, for example, excessive swapping.
This release includes support to .NET for the Globals API. This enables direct global access and manipulation of globals from .NET. This API enables .NET and Caché to run intraprocess on the same computer providing high speed data access.
This version of Caché allows XEP to run over TCP. XEP is a lightweight Java object persistence technology for Caché. Initially it executed intraprocess because it was built only on top of the Globals API for Java. While it still leverages the Globals API concepts in many respects, being able to use TCP means that it is now optionally intraprocess.
.NET 4.0 Components For Object Provider And .NET Gateway
This release provides .NET 4.0 compiled versions of the Object Provider for Caché and the .NET Gateway. Though previous releases were compatible with .NET 4.0, there is now an option to only have .NET 4.0 components in applications.
Long Strings Are The Default
Beginning with 2012.2, support for long strings is now enabled by default for new installations as well as upgrades.
Performance And Scalability
Beginning in this release, rollback operations issued by a process on an ECP Application Server will be addressed by the Database Server asynchronously. This approach allows continued parallel processing of other requests. Previously, a rollback operation was handled synchronously: the Database Server would serially handle the entire rollback before continuing with other requests from a given ECP Application Server. The new technique prevents a long rollback operation from disrupting other application work on a given ECP Application Server.
Use AES Hardware Encryption When Available
Many processor vendors now include support for encryption algorithms directly in hardware. Beginning with this release, Caché takes advantage of this feature (when present) in the Advanced Encryption Standard (AES) calculations. In doing so, it provides increased performance by using the instruction set of the processor.
The initial use of this option will take place on Intel 64-bit processors, beginning with the Intel® Xeon® Processor X5680 (Westmere). On such systems, Caché makes direct use of the hardware instructions to perform AES encryption. The systems for which this is true are: Microsoft x86-64, Red Hat Enterprise Linux for x86-64, SuSE Linux Enterprise Server for x86-64.
Reliability, Availability, Maintainability, Monitoring
Simpler Disaster Recovery Failover And Failback
In previous releases, Caché Database Mirroring has provided an Asynchronus Mirror configuration for Disaster Recovery purposes. In this release, InterSystems has made it easier to switch over to an Asynchronous DR Mirror node in the event of a disaster (or for testing purposes); and also made it easier to switch back to the main production failover mirror at a later date.
Configurable UNIX® Installation
In this release UNIX® installations can now be done in a manner similar to Windows installs that use command line options. New configuration parameters are available with the new cinstall_silent
installation script to define the instance name, installation directory and packages to install.
Public Key Infrastructure
Increased concerns about information security in many industries have correspondingly increased the need to encrypt and sign documents using the X.509 certification. However, generating, safe-guarding, and distributing X.509 keys and certificates is a complex and time-consuming process.
In this release, InterSystems users have the ability to automate the process that generates certificates by defining a server to act as a certificate authority. Clients cannot only request certificates from the certificate authority, but automatically receive them for their local use.