Presenter: Mark Bolinsky
Task: Provide failover for distributed systems without using a VIP
Approach: Demonstrate using InterSystems’ database mirroring with external traffic managers such as F5 LTM/GTM
 

With distributed environments and even public cloud environments, the use of a VIP sometimes is not desirable or even possible given network topology or deployment. The session will demonstrate integrating database mirroring with external traffic managers such F5 LTM/GTM using API based triggers in InterSystems products to interface with the F5 appliances. This not only presents automated redirection for the local mirror members, but also provided automated client redirection to asynchronous DR mirror members. 

 

Content related to this session, including slides, video and additional learning content can be found here.

00
0 0 249

Presenter: Mark Bolinsky
Task: Decide whether a converged infrastructure is ideal for your enterprise applications
Approach: Discuss best practices and provide guidance on the right questions to ask
 

The traditional use of “SAN storage” is no longer the only choice for deploying enterprise application. Software defined data centers are making inroads into enterprise data centers, and there is good reason for it. There is the potential for significant infrastructure cost savings, architecture simplification, reduced administration costs, and depending on the configuration - even better performance. This session will discuss some best practices and outline decision guidance to help you ask the right questions when considering hyper-converged architectures.

 

Content related to this session, including slides, video and additional learning content can be found here.

00
0 0 208

++ Update: August 1, 2018

The use of the InterSystems Virtual IP (VIP) address built-in to Caché database mirroring has certain limitations. In particular, it can only be used when mirror members reside the same network subnet. When multiple data centers are used, network subnets are not often “stretched” beyond the physical data center due to added network complexity (more detailed discussion here). For similar reasons, Virtual IP is often not usable when the database is hosted in the cloud.

Network traffic management appliances such as load balancers (physical or virtual) can be used to achieve the same level of transparency, presenting a single address to the client applications or devices. The network traffic manager automatically redirects clients to the current mirror primary’s real IP address. The automation is intended to meet the needs of both HA failover and DR promotion following a disaster. 

130
1 12 3,594
Article
Mark Bolinsky · Feb 2, 2016 1m read
Cache' databases as UNIX sparse files

Some third party backup products may by default restore CACHE.DAT files as UNIX sparse files when there are trailing zeroes in the backup file.

The support for sparse files vary from UNIX distribution and file system types.  Sparse files attempt to use file system space more efficiently when blocks allocated to the file are mostly empty similar to thin-provisioned storage.  The file system transparently converts metadata representing empty blocks into "real" blocks filled with zero bytes at runtime. The application is suppose to be unaware of this conversion.

50
0 0 261

** Revised Feb-12, 2018

While this article is about InterSystems IRIS, it also applies to Caché, Ensemble, and HealthShare distributions.

Introduction

Memory is managed in pages.  The default page size is 4KB on Linux systems.  Red Hat Enterprise Linux 6, SUSE Linux Enterprise Server 11, and Oracle Linux 6 introduced a method to provide an increased page size in 2MB or 1GB sizes depending on system configuration know as HugePages.

At first HugePages required to be assigned at boot time, and if not managed or calculated appropriately could result in wasted resources.  As a result various Linux distributions introduced Transparent HugePages with the 2.6.38 kernel as enabled by default.  This was meant as a means to automate creating, managing, and using HugePages.  Prior kernel versions may have this feature as well however may not be marked as [always] and potentially set to [madvise].  

Transparent Huge Pages (THP) is a Linux memory management system that reduces the overhead of Translation Lookaside Buffer (TLB) lookups on machines with large amounts of memory by using larger memory pages.  However in current Linux releases THP can only map individual process heap and stack space.

40
2 8 3,407

The release of IBM POWER 8 processors with AIX 7.1 introduced up to 8 SMT threads per processor core (logical or physical).  Which SMT level (1, 2, 4, or 8) to use can be confusing and varies based on multiple factors.  This article is meant to help with a starting point for your specific application.

Firstly, if running on a version of 2014.x or older, it is advised to use SMT 4 or lower.  SMT 8 with those older versions of Cache' has shown a decline in performance and scaling in benchmarking applications.

30
0 3 2,163

Often times support and sales engineers are asked about recent benchmark results on various platforms and large scale configurations.  These will be made available here in the Developer Community in the "Documentation" section, and as an example here's a link to a recent Intel E7 v2 series processor benchmark.

https://community.intersystems.com/documentation/data-scalability-intersystems-caché-and-intel-processors-0

00
0 0 192
Article
Mark Bolinsky · Nov 19, 2015 1m read
Disk and storage design considerations

There are many storage technologies available today from various vendors.  The storage technology and configuration best for your application depends on the application access patterns and workloads.  

The attached document discusses the various design considerations and recommendations for various technologies.  This guide is to help you during discussions with your storage vendor to determine the appropriate storage technologies and products that will work best to meet the performance goals for your applications.

70
0 3 388