<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Igor Nemy on Medium]]></title>
        <description><![CDATA[Stories by Igor Nemy on Medium]]></description>
        <link>https://medium.com/@igornemy?source=rss-7e30ed539729------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>Stories by Igor Nemy on Medium</title>
            <link>https://medium.com/@igornemy?source=rss-7e30ed539729------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 23:32:58 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@igornemy/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[What exactly RPO means]]></title>
            <link>https://igornemy.medium.com/what-exactly-rpo-mean-48c31e9c1c80?source=rss-7e30ed539729------2</link>
            <guid isPermaLink="false">https://medium.com/p/48c31e9c1c80</guid>
            <category><![CDATA[rpo]]></category>
            <category><![CDATA[disaster-recovery]]></category>
            <category><![CDATA[design]]></category>
            <dc:creator><![CDATA[Igor Nemy]]></dc:creator>
            <pubDate>Mon, 29 Jun 2015 16:37:42 GMT</pubDate>
            <atom:updated>2024-01-07T13:13:35.935Z</atom:updated>
            <content:encoded><![CDATA[<p>During any Disaster Recovery planning you probably operate such SLA terms like RTO and RPO. They are very common and extremely important notions which can simplify research of right method/tool to meet your DR requirements. In short, these abbreviations stand for Recovery Time Objective and Recovery Point Objective which you of course know. But, I found that far not every one specialist who participates DR negotiations fully understand what exactly RPO means and how it is actually performs.</p><p>So, in short, RPO is measured in time and indicates a portion of data that you tolerate to lose during a failure or disaster. Most people think that if set replication or backup job frequency to, as example, 1 hour, they will get RPO of 1 hour. But unfortunately this is MISLEADING thought! Actual RPO will be the time to transfer data change which might be much longer than 1 hour. You may ask, WHY? Ok, I will explain on the given example.</p><p>Replication/backup job is set to 1 hour. Our dataset, i.e. combination of data blocks (or any other granular objects), is constantly changing over the time. Not fast, but it changes. And our business requirement is to be able to recover data younger than given 1 hour interval. Each 1 hour interval! For each of data block! See? No? Ok, read further.</p><p>Here is illustration. For example, our given data block was committed at 10am and replication job took up to 50 minutes to to transfer this point to recovery site. Thus at 10:50am we have recovery point with condition of 10am and it will have only 10 minutes before it will be stale and don’t much our given RPO rule. So, we must immediately perform another replication job with lightning speed to fully transfer changes of last 50 minutes of that given block to recovery site in next 10 minutes. And no longer! If transfer for next replication job performs longer that 10 minutes, for example 40 minutes, we will have RPO violation between 11:00am and 11:30am, because first recovery point will stale at 11:00am and second recovery point will be received only at 11:30am.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/694/1*1nikKmb9Pkl-fipxMHPi7Q.png" /><figcaption>RPO timeline</figcaption></figure><p>So, why I must to know actually how often replications perform if replication algorithm deals with this, you may ask? Answer as usually lies in proper planning of DR design, and in this case — in the network saturation consideration to fully comply with replication requirement on amount of channel bandwidth between protected and recovery sites.</p><p>BTW, ask your vendor representative to clarify how exactly their replication tool deals with RPO in your environment.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=48c31e9c1c80" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Large Multi-Site Disaster Recovery Protection Topology]]></title>
            <link>https://igornemy.medium.com/large-multi-site-disaster-recovery-protection-topology-c998407e3b82?source=rss-7e30ed539729------2</link>
            <guid isPermaLink="false">https://medium.com/p/c998407e3b82</guid>
            <category><![CDATA[multisite]]></category>
            <category><![CDATA[disaster-recovery]]></category>
            <category><![CDATA[global]]></category>
            <dc:creator><![CDATA[Igor Nemy]]></dc:creator>
            <pubDate>Sat, 04 Apr 2015 22:03:30 GMT</pubDate>
            <atom:updated>2024-01-07T08:41:13.550Z</atom:updated>
            <content:encoded><![CDATA[<p>Disaster recovery solution for virtual infrastructures is mandatory consideration during planning design of any modern data center nowadays. But if for simple model of the protection of one large data center to remote recovery site is everything pretty clear, in cases of large topologies with a lot of distributed local branches across territorial units such as states, countries or even continents, the multi-site protection may be a quite interesting challenge.</p><p>Just in case you don’t read <a href="https://medium.com/@inemylostyvyi/comparison-nutanix-versus-vmware-vsan-f701581b884">my previous article</a>, where I have already a little bit touched upon disaster recovery topic, I would to remind you that:</p><blockquote>In case of the loss of the entire physical site due to natural disasters, technogenic catastrophes or other force majeure, data protection via replication to other availability zones should be considered and accurately designed. Data protections can be designed either with completely zero RTO and RPO or with strictly defined data loss acceptance and clear predictable recovery time. Speaking about storage platforms nowadays, first (zero RPO) can be achieved via synchronous replication and second via asynchronous, when written data arrives to remote site with a little delay, during which data delta loss is acceptable.</blockquote><p>For topologies with a lot of branches the design of business continuity solution can lead the architect’s thoughts to these key ideas:</p><ul><li>instead of using one global recovery site (what even sometimes is impossible due to latency or etc.) the alternative way of using many distributed decentralized recovery sites may be considered, when each acts as a regional recovery site for its closely located protected sites in a “bunch”;</li><li>since protected resources in a “bunch” are distributed across territorial branches the likelihood of simultaneously failed branches is tends to zero and thus amount of normally idle compute resources for recovery site may be reduced to an equivalent of only one protected branch;</li><li>optimizations such as deduplication and compression capabilities may be considered for recovery sites to achieve reduction for storage capacity resources;</li><li>and my favorite point—we can declare ultra critical level for services by protecting them twice across two different regions: one to local recovery site inside local “bunch” and one to another nearest “bunch”. These replications should be separated and due to using connectivity with different latencies should support independent RPO calculations.</li></ul><p>To include these core ideas during building appropriate disaster recovery solution, one of the possible multi-site protection topologies can be a circular protection:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*n3zDpI2o-P7L4-5c_C6xjQ.png" /><figcaption>Multi-site Disaster Recovery Circular Protection Topology</figcaption></figure><p>Speaking about virtual infrastructures nowadays, to build whole or just parts of proposed above topology the final design can leverage a lot of various products available on the market. Let me show you a few specifications below as an examples:</p><ul><li>As a virtualization platform is used vSphere and as a storage is used Nutanix SDS with asynchronous replication within a bunch. As a disaster recovery automation solution can be used VMware Site Recovery Manager (SRM). And to make one additional replication to nearest bunch for ultra critical services is used vSphere Replication with further manual recoveries, since SRM prevents protect the same VM twice;</li><li>For the cases when vSphere and Hyper-V hypervisors are used in neighboring bunches, as a replication and disaster recovery automation might be used Zerto Virtual Replication.</li></ul><p>I would like to say that over next 2–3 years all these benefits of all possible disaster recovery topologies will be implemented in leading SDS products.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c998407e3b82" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Software Defined Storage (SDS) vs Traditional SAN]]></title>
            <link>https://igornemy.medium.com/software-defined-storage-vs-traditional-san-157c3a71be7f?source=rss-7e30ed539729------2</link>
            <guid isPermaLink="false">https://medium.com/p/157c3a71be7f</guid>
            <category><![CDATA[compassion]]></category>
            <category><![CDATA[software]]></category>
            <category><![CDATA[cloud-computing]]></category>
            <dc:creator><![CDATA[Igor Nemy]]></dc:creator>
            <pubDate>Sat, 14 Feb 2015 13:32:32 GMT</pubDate>
            <atom:updated>2024-01-07T08:18:08.810Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/982/1*7ieVFzPS4Ne1VNygVAZ78g.png" /><figcaption>SDS vs traditional SAN</figcaption></figure><p>This week I had a couple hours to analyze about almost everything what I knew about the Software Defined Storage (SDS) paradigm. And as a result of these reflections I have defined next main advantages of the SDS against old traditional SAN at the current moment:</p><ol><li><strong>Deep SLA Atomic Granularity</strong>. Since SDS brings a new level of completely separated logical storage service from its hardware, we can achieve deeper granularity of stored elements for which we enable such features as rebalancing, migration, replication, deduplication, snapshotting and so on. All these tasks might be simply configured on per unit basis to achieve maximum SLA uniqueness. For example, instead of whole LUN or Volume aggregation tuning, we can create unique SLA policy for particular virtual disk. This also empowers us to make data portable, eliminating planned and unplanned downtime on per VM or per virtual disk basis across different physical disks, nodes, racks, rooms and so on.</li><li><strong>Pay As You Grow model. </strong>Because SDS does not have any central elements, we will not have central bottlenecks or single point of failures and can have theoretically unlimited horizontal scalability range and vertical range is limited only to servers hardware limitations here. And since we can scale (or modernize) our storage cluster on per drive basis, we can achieve cool approach of very evenly investment during building our storage platform. This is something like Pay As You Grow model ☺</li><li><strong>Very low OPEX. </strong>In traditional old paradigm, you would need to plan the lifecycle of the future storage system. During planning stage, you need to understand how much disk space and how much I/O commands it should serve and at the same time taking into account how it can operate, and how it scales. You also should to consider whether it requires planned downtimes to implement new functionalities or disabling some extra ones. Or how to implement current or future disaster replication and so on. After understanding future storage architectures of this life cycle, you must invest in building SAN or modify your existing one. Thus, if you chose to present expanded capacity to a new compute server, you will require not only adding new disks to storage and creating new LUN, but there will be necessity to perform separate fabric zoning and multipathing driver installation. <br>In the new software defined paradigm you can find completely new operations approach — no RAID related calculation, no SAN setup, no zones creation, no special cabling or special switch hardware configuration. For SDS the disks, nodes, and rooms, are all suitable duplication locations. You can easily add new disks to nodes, or node to rack, or rack of nodes to a system without any downtime. Rebalancing, migration, new replication and so on can be just programmed because this storage is already the program.</li><li><strong>Investment Protection. </strong>In SDS, as in any other Software Defined things, subject to obsolete is mostly the software layer. Meanwhile used hardware is simple commodity elements such as disk drives, memory, CPU and so on, all “magic” happens inside of the logical subsystems, implemented completely in the software layer. Thus, to implement any new feature you need only to update the software layer. This is great protection of investments.</li></ol><p>If you, my dear reader, see any other obvious distinctions, please leave comment here or over email, I will very appreciate this.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=157c3a71be7f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What to expect in 2015 for cloud/virtualization IT area]]></title>
            <link>https://igornemy.medium.com/what-to-expect-in-2015-for-cloud-virtualization-it-area-197587231e6b?source=rss-7e30ed539729------2</link>
            <guid isPermaLink="false">https://medium.com/p/197587231e6b</guid>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[predictions]]></category>
            <category><![CDATA[opinion]]></category>
            <dc:creator><![CDATA[Igor Nemy]]></dc:creator>
            <pubDate>Sat, 03 Jan 2015 10:36:23 GMT</pubDate>
            <atom:updated>2024-01-07T08:00:40.297Z</atom:updated>
            <content:encoded><![CDATA[<p>At last week of 2014 year I spent few hours reading very interesting researchers, interviews and predictions materials from major IT leaders, market analyst companies and field gurus to better understand what to expect in new 2015 year for us, IT experts in cloud/virtualization IT area.</p><p>So here is a short top of top lists of the major IT trends in cloud/virtualization market:</p><ul><li>This year will be a year of growing <a href="https://dzone.com/articles/why-you-need-move-your-data">Software Defined based infrastructures</a>. Especially we will see many implementations of Software Defined Network and Software Defined Storage deployments. Key players around of SDN and NFV concepts expect huge consolidation and in 2015 we will see clear rank of them with a top player, and as result, strong frozen vector of development for these concepts. In its turn, SDS due to its capabilities of shared nothing, flexible multilevel duplications and usage of commodity hardware, will finally overtake traditional SAN paradigm.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RV8-qp1sqp0DtAXLlNGCCw.png" /><figcaption><em>Traditional data center vs Software Defined</em></figcaption></figure><ul><li>Hybrid Clouds/Cloud Federation — this is key model that we will see across customers interests and readiness. Due to rising interest to SaaS assets against the background of ending support for windows 2003 (July, 2015), IT departments will review theirs on-premise and SaaS assets balance. Software defined approaches will transform private/public usage model to hybrid usage model.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8xt3dBJ80Xb1-L3zF8ZbsA.png" /><figcaption><em>Software Defined Infrastructure</em></figcaption></figure><ul><li>Also this year will be a year of flash storage devices. Flash memory industry comes closer to that cost-effective barrier, when storing data on the only flash devices in additional to proper management of this data is finally economically appropriate.</li><li>Containers. Such things as Docker, Project Atomic, Warden (Garden) with Diego and <a href="http://kubernetes.io/">Kubernetes</a> orchestration in Cloud Foundry and OpenShift PaaS make new concept of remixing exists and future IT assets. Containers will be hot topic in this year due to nature of their easiness in deliveries and operating deployment activities.</li><li>Data Privacy Officer role will be one of the most important roles for enterprises. This means that either we will wear this role to ourselves or it should be next hire at your organization.</li></ul><p>So, please enjoy Happy New Year 2015! Stay involved! Peace to all!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=197587231e6b" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>