Solution Portfolio

get insights on our solution portfolio offerings

HiPAS: High Performance Adaptive Schema Migration

This tool is designed for a fully automated transfer of schemas between Oracle databases. For this, the source schema is analyzed and the transmission is controlled and parallelized by an optimizer.

The performance data of the source and target systems are continuously monitored, so that the optimization of the transfer of the schemas is always automatically adjusted to the systems’ needs.

If the data transfer is supposed to load the source or target system only up to certain limits, the limiting values will be communicated to the optimizer which will then control the migration of data within the specified limits. Running the systems with better utilization during the night or weekend is also easily manageable. Character set conversions of any type are possible as well. RAC is supported, and aims to increase the transmission performance.There is no need for temporary data storage.

Advantage: easy operation, high performance achieved with parallelism, supports quality of service, character set conversion, no buffering needed. A Mimium Downtime Option is avalaible which allows the source database to stay up and running while data is transfered in the background. All changes applied to the source by various applications will be captured and also replicated by HiPAS. This allows 20+ TB database systems to be logically migrated within downtime windows of 2 or 3 hours.

The tool is suitable for databases of Oracle Enterprise Edition and Oracle Standard Edition.

LoTTA: Load testing & test automation framework

Determining the load limits and significant predictions of performance data for highly complex integrated systems in the database area is a difficult and possibly lengthy endeavor. Setting up realistic tests based on existing applications is expensive, complex, and often a time-consuming process.

At this point, LoTTA comes into play. LoTTA is a THIN client, which can generate highest database workloads with a default client within predetermined limits. For this purpose it can be exactly parameterized in a simple way how the target system should be loaded in relation to transactions, CPU and I/O.

The load characteristics, in particular, can be differentiated very accurately, e.g. the I/O load split up in random I/O and sequential I/O can be determined, etc. The tool can store the performance and analysis data in a repository database to keep it available for later evaluation or comparison. In this way, performance data can be compared across many hardware generations.

The tests can be fully automated and, with minor adjustments, automated failure tests are also possible. For documentation purposes the log files can also be loaded automatically into the repository database. The complete data can be edited and queried in the repository database using SQL.

PeLA: Performance –and Lock Analysis for Oracle SE

The "post-mortem" analysis of SQL problems in the Oracle Standard Edition is severely limited. The upgrade to Enterprise Edition to improve the analysis possibilities is, due to cost reasons, only reserved for a few customers. This is the point where Pela starts to work. Pela collects, in addition to Statspack (Oracle product), key performance data on session level for later analysis.

Thus, problems of individual sessions can be analyzed, investigated and solved for the future, even afterwards.

IOMeshine

I/O Meshine is an open, scale out storage architecture for Linux, Oracle, Solaris and Windows - developed by the pasolfora storage team - that is flash only, supports file and block based access end common connectivity technologies like Infiniband , 1 Gbit and 10 Gbit Ethernet, 8 and 16 Gbit Fibrechannel, iSCSI, FCoE, CIFS, NFS, etc.

Linux applications i.e. databases can also be run directly on the storage server (up to 2 TB of main memory available)

A small configuration comes in 19 inch 4U rack mount format, provides approx. 10 TB of usable storage and has proved to deliver

  more than 300.000 Read IOPS
  more than 5 GB sustained Read Rate per second
  more than 50.000 Writes IOPS
  more than 750 MB per second oracle redo log write rate

The rates for Redo and Data Write are sustained rates that have been measured after hours of production workloads and after having already hit the flash write cliff and suffering from the flash erase process (garbage collection). The write rates before hitting the write cliff are significantly higher.

I/O Meshines only consist of standard, off-the-shelf x86 server components and SSDs, open source software and german engineering. Therefore it is fully flexible and adjustable to your needs and can easily be integrated in your inhouse technologies and products i.e. Veritas, ScaleIO, vSAN, Oracle ASM, RAC, Data Guard and Flashback.

AMon: Archive Monitor for Oracle

Archive Monitor for Oracle is a highly customizable software framework to optimize rman backups against any type of rman compatible tape or disk device. It is designed to handle large workloads in previously defined backupset sizes. The software provides the ability to save backupsets of any size to any number of backupdestination. Hereby it provides the possibility to control the parallelism of active backupsets.

Furthermore Archive Monitor for Oracle provides an internal status table, which monitors the processing of backups and provides restart capability in case of an unexpected abortion. Due to this internal logging, it is nearly impossible to delete archivelogs not yet saved to all configured backup destinations. Archive Monitor for Oracle can be included in the Oracle GRID Infrastructure as an autonomous cluster ressource or simply be put to work by the included init script. The complete Softwarestack is deployed once per operating system / server. Adding Databases is done by creating a directory structure with the included setup script and configuring the needed parameters for the database.

Archive Monitor for Oracle is RAC aware and after installation and configuration able to run on any clusternode with an active database instance. Configuration files are automatically transferred to all clusternodes, and the newest configuration is taken as the active configuration.

In case of a RAC database, only one instance of Archive Monitor can run at the same time. In this case it is recommended to use the installed clusterware to take care of this. In addition to these standard features there is the possibility to backup a set of archivelogs in a priority saveset. This set of archivelogs can be defined by two SCNs for example. This functionality can be used to connect Archive Monitor for Oracle to an online database snapshot tool e.g. for cloning databases. Archive Monitor for Oracle can easily be adopted to the customers needs due to a very generic function programming respectively object orientation.

Key Features:

  • arbitrary number of backup destinations
  • configurable amount of archivelogs per backup job
  • parallel save of a backup job to all configured destinations
  • parallelism of concurrent backup jobs configurable
  • immediate deletion of all archivelogs in a backup job after successful save to all backup destinations
  • validation of successful backup by inspecting several oracle and rman views in combination with the internal status table
  • restartable after program errors
  • can be included in oracle clusterware as a ressource
  • can be used on RAC databases
  • housekeeping for internal tables and temporary files
  • internal status table
  • usage of standard rman commands – use of all other rman tools for restoring or checking backups possible
  • work around oracle bugs while saving archivelogs by inspecting multiple views and logs.
  • Interface for priority save of logs between two SCNs for connecting other backup or clone tools.
  • Support for multiple databases per host.
  • Init script for usage without clusterware
  • setup script for simple deployment and configuration
  • fully customizable to suit your needs

PDA - Pasolfora Database Appliance

We provide a fully automated solution for Linux and Oracle RAC deploy and upgrade processes (even major release upgrades). We even have loadtesting capabilities to detect significant workload anomalies before and after changes compared to a given baseline.

Available as Hardware Bundle and Software only.