Distributed operating system: Difference between revisions

Content deleted Content added
JLSjr (talk | contribs)
No edit summary
JLSjr (talk | contribs)
No edit summary
Line 33:
<br /><br />
}}
{| class="wikitable" cellpadding="0" style="border:1.0px solid black;color:#000;margin-left:18em;"
|-
|style="background:#FDEE00;"|[[Image:Nuvola apps korganizer.svg|35px]]
|style="background:#DFFFFF;width:500px;"|'''Rebuilt lead, in preparation of "Architectural features" overhaul'''<br />&nbsp;&nbsp;&nbsp;[[User:JLSjr|JLSjr]] ([[User talk:JLSjr|talk]]) 10:37, 7 April 2010 (UTC)
|}
{| class="wikitable" cellpadding="0" style="border:1.0px solid black;color:#000;margin-left:18em;"
|-
Line 86 ⟶ 91:
<hr style="width: 80%; height: 2px;">
 
<!-- A '''Distributed operating system''' is the minimal subset of software within a distributed system, which -- considered collectively -- provide all operating system services required to support higher-level components in empowering and maintaining the system. -->
 
A '''Distributed operating system''' is the composite aggregate of operating system software in a Distributed System. This software – considered collectively – supports the concerted operation of the system’s independent, autonomous, and disseminated computational units. Each discrete unit contains an allocation of operating system software, which generally consists of two distinct portions. One portion, the Kernel, is a minimal subset of system software – usually common to all units – devoted to the control of a unit’s underlying hardware resources and devices; it also supports the unit’s remaining portion of system software. A unit’s remaining system software, the local Operating System, is an ad-hoc collection of discrete system software components, formulated to empower the unit’s operation. The kernel and the local operating system work collaboratively to ensure each unit's cooperative contribution to the whole System. This collective cooperation, sustained at the operating system level, gives the distributed system its unified appearance.
{{Userbox
|border-c=#000
|border-s=1
|id-c=LightSteelBlue
|id-s=12
|id-fc=#000
|info-c=#fff
|info-s=8
|info-fc=#000
|id=[[Image:Nuvola apps kchart.svg|40px]]
|info=A Diagram will be furnished to assist in illustration of this idea.
|float = right}}
As a minimal software composition, a distributed operating system Kernel is often referred to as a Microkernel. The minimal nature of the kernel strongly enhances modular potential. The kernels ubiquitous quality also supports greater opportunity for system flexibility and scalability. Minimal by design, the microkernel usually contains only mechanisms and services which otherwise removed, would render the system incapable. The microkernel primarily provides lower-level resource, process, communication, and I/O management. These services are made possible by the exposure of a comprehensive, yet concise array of primitive mechanisms. Through these primitives, the microkernel supports the higher-level functionality of its local operating system. This clear distinction between the microkernel and local operating system offers a clean separation of mechanism and policy; the difference between what is done, and how or why it is done, respectively.
 
A unit’s operating system is primarily characterized by its composite nature. This system software composition is dictated mainly by the unit’s responsibilities to the overall system. These responsibilities focus principally on the allocation, management, and disposition of system processes and resources in fulfillment of global system policy. The unit operating system participates largely through the kernel’s communication services in fulfillment of its system-wide responsibilities. This multi-level collaboration between a kernel and a unit operating system, and in turn between the units in a distributed system, is the key function of the distributed operating system. However, this function comes at a very high price.
<br />
{{Userbox
|border-c=#000
|border-s=1
|id-c=LightSteelBlue
|id-s=12
|id-fc=#000
|info-c=#fff
|info-s=8
|info-fc=#000
|id=[[Image:Nuvola apps kchart.svg|40px]]
|info=A Diagram will be furnished to assist in illustration of this idea.
|float = right}}
The logical price of distributed system – including its operating system -- is calculated in terms of overcoming vast amounts of complexity on many levels, and in many areas. This calculation also includes the depth and breadth of design investment and architectural planning required in achieving even modest levels of success. These development considerations are critical and unforgiving in that an overwhelming majority of a distributed system’s architectural, design, and implementation details are required from the start. A huge amount of research work has been done in the past towards distributed computing, and continues today.
 
While the research and implementation efforts came to initial commercial success in the late ‘70s through the mid-‘80s, the distributed system has a much richer historical perspective. There are several instances of interesting, innovative, and almost radical implementations of distributed system components dating back to the mid-‘50s. The emergence and proliferation of multiprocessor systems and multi-core processors has led to a re-emergence of the distributed system, and an enormous increase in quality research. Many of these research papers describe plausible paradigms for the future of distributed computing.
 
== Description ==