Video
with the manuscript function you'll be able to create your own digital lecture manuscript.
However, in order to link all your notes with your user profile it is required that you
login to the tele-TASK portal to use this functionality.
If you don't have an account yet, you may register for a tele-TASK account here.
with the marker function you'll be able to create your own digital time markers.
However, in order to link all your markers with your user profile it is required that you
login to the tele-TASK portal to use this functionality.
If you don't have an account yet, you may register for a tele-TASK account here.
Please enable javascript to use this function.
Dear user,with the tagging function you'll be able to add taggs to videos.
However, in order to link all your tags with your user profile it is required that you
login to the tele-TASK portal to use this functionality.
If you don't have an account yet, you may register for a tele-TASK account here.
No tags have been added to this content so far.
Tags added to the series of this content:algorithmen (1); backup (1); Datenbanken (1); dbms (1); festplatte (1); performance (1); Persistenz (1); recovery (1); Speicherhierarchie (1); ssd (1); test (1); vorlesung (1);
Please enable javascript to use this function.
Dear user,with the links function you'll be able to add links to other resources to this content.
However, in order to link all your links with your user profile it is required that you
login to the tele-TASK portal to use this functionality.
If you don't have an account yet, you may register for a tele-TASK account here.
No links have been added to this content so far.
with the playlist function you'll be able to create your own lecture video playlists.
However, in order to link all your playlists with your user profile it is required that you
login to the tele-TASK portal to use this functionality.
If you don't have an account yet, you may register for a tele-TASK account here.
This content is not used in any playlist.
- Overview (00:00:01)
- CO lSe lSLIS Protocols (00:00:44)
- Consensus (00:01:36)
- DiStFibLJtEC COITIITWE PFOt0C0 S (00:03:44)
- with network part tions (00:03:44)
- (commit or abort) (00:03:44)
- Requirements for the distributed case (00:03:44)
- Two Phase Commit (ZPC) (00:07:22)
- ZPC (00:09:36)
- ZPCZ SOITIE PFOb 6ITlS (00:14:08)
- Three Phase Commit (3PC) (00:17:18)
- Three Phase COlTllTlit( (3P ) (00:19:32)
- Three Phase Commit (3PC) (00:21:16)
- ZPCZ SOITIG Pl Ob emS (00:21:49)
- Three Phase Commit ( (3PC) (00:21:50)
- Overview (00:22:03)
- OLTP SY5tEI T1SI H StOl 6 (00:22:03)
- Distributed Key Value (00:22:03)
- Distributed Data Stores (00:22:03)
- WDM for operat onal dambases (00:22:03)
- WDM for analytic databases (00:22:03)
- Three Phase Commit (3PC) (00:22:04)
- acknowledged Prepare Tof ommit wlth Ready To Commit (00:22:04)
- Workers which have rece vec( P? are To ComrnTF know (00:22:04)
- fis permanently down workers will block inde nitely (00:22:04)
- Three Phase Commit( (00:22:23)
- Three Phase Commit( (3Pc) (00:22:30)
- Overview (00:23:15)
- Three Phase Commit (3PC) (00:23:16)
- Overview (00:23:34)
- Three Phase Commit( (00:23:39)
- Three Phase Commit (BPC) (00:23:40)
- 2PC SOIT16 PFOb 6IT1S (00:23:41)
- Two Phase Comrnit (ZPC) (00:23:42)
- Distributed COITHTHE PI OCOC0 S (00:23:44)
- Overview (00:23:45)
- COFISEFISLIS (00:24:24)
- Terrr ination all nodes eventually decide (00:24:24)
- D ff cult in the presence of failur imposs ble to solve in certain (00:24:24)
- Naive approach (assumption no fau t s) Each node sends vote to all (00:24:24)
- Two Phase Commit (ZPC) (00:24:25)
- W det mbuted databases and elsewhere (00:24:25)
- @ global log (00:24:25)
- one coordinator (arb trary site) (00:24:25)
- Important consensus problem for distributed transactions (00:24:25)
- ZPC (00:24:27)
- 2PC Some Problems (00:24:28)
- Three Phase Commit (3F Cf (00:24:29)
- Overview (00:24:30)
- Distributed Hash Tables ( (DHT) (00:25:25)
- Specific systems differ in their topoiogy (00:25:25)
- Data centered routing without global knowledge (routing by value) (00:25:25)
- Partitioning of the logical key space among all peers (00:25:25)
- get(key) vaiuez retrieves the value associated with the (00:25:25)
- put(key value) stores the pair (key value) in the hash table (00:25:25)
- Hash table providing two operations (00:25:25)
- DHT Design Goals (00:31:48)
- Ch0rCl (00:32:30)
- For N peers O( og N) entries (00:32:30)
- Mapping of keys to nodes which maintain the corresponding key (00:32:30)
- Chord Data Organization (00:35:07)
- Chorcl Identifier Circle (00:36:27)
- Chorcl Naive Search (00:37:34)
- Ch0rCl Scalable Search (00:38:40)
- Finger Table Example (00:39:35)
- Chor tlz Scalable Search (00:40:37)
- Finger Table Example (00:40:50)
- Scalable Search 2 (00:41:13)
- Chord Node Joins and Ring V liI 1te 1 I 1C9 (00:44:18)
- Check and update all finger (00:44:18)
- are up to date (00:44:18)
- Beware not all structures (00:44:18)
- Redistribute keys (00:44:18)
- Chorcl Nocle Joins ancl Ring V il ltel 1 l lCe (00:45:43)
- Content Acldressable Netv orl (CAN) (00:46:52)
- responsible for Its own zone (00:46:52)
- u Mapping of d dimensional Cartesian coordinate space to d (00:46:52)
- Ü Detefmine SUCCSSSOF FIOÜG (00:46:54)
- Content Adclressable Network (CAN) (00:48:51)
- Content Aclclressable Network (CAN) (00:51:50)
- Content Aclclressable Network (1 (CAN) (00:51:51)
- Routing (00:52:27)
- Routing Example (00:52:28)
- CAN Data Organization (00:53:11)
- Content Aclclressable Network (CAN) (00:53:12)
- responsible for its own zone (00:53:12)
- Partitioning coordinate space in a way that each node is (00:53:12)
- Mapping of d dimensional Cartesian coordinate space to d (00:53:12)
- CAN Data Organization (00:53:27)
- base operation ookup(k) (00:53:27)
- (k v) is stored at node responsible for zone Z that contains p (00:53:27)
- Rouüng (00:54:02)
- Rouüng (00:55:39)
- Routing Example (00:57:50)
- Routing Costs (00:57:51)
- Split Example (00:57:52)
- Rouüng Costs (00:58:57)
- CAN Node Joins (00:59:03)
- Split Example (00:59:33)
- Leaving the CAN (00:59:44)
- else respond with own TAKEOVER message (01:00:54)
- On timer expiration send a TAKEOVER messageto all (01:00:54)
- Overview (01:03:10)
- atform development (01:04:44)
- Sources for Slides (heavily used) (01:07:06)
- Overview (01:07:36)
- Cloucly weather (01:07:39)
- What is the CIOUCI? (01:09:08)
- It s comp ete gibberish It s insane When ws kh diocy going to stop? (01:09:08)
- details are abstracted from the users who no longer neecßwf V OFD (01:09:08)
- NIST has put together E characteristics of cloud computing to help define it (01:10:05)
- e a we W em e se e s e sa se ms ea W within mvndnee and men once wmf demand is gone wu san sca e back dawn cd ene hundred servers measured service That Clouds enable a mecennq of sen sg t exng used fd admmaneawy ednwd anddpurm e me resource use And typicaliy thatis often ddne mrdddn Pay Per use where you have to pay forserver ame sd vddwe ner gd ng cd mg au me servers in me sviem 1 knaw ma me agency bu t essen a ya private m ema doud and mey wanted to do me Hd y Gra df exddd edmpmng sd mefease serverutii zat on from ten percentto e ghtv percent wnam s agency found xsthat smce mere was nd rneasurvngdfme sen ee ur op miz ng usage by me business users me agency was Just using everything man wasava ame 50 their s rver u lila went uD (0 ighty Gertent butthev did f ackua y free UD any resources (01:10:05)
- mwu = y me away W away e a Ww ee am La eu eme W ae I ve e a a ea abmw mars being prwaea So vr not Ws sea api iry Scalabmty vom abww tu increase mearch tectureto aüow fer mare demand ßucme idea lw th Rau d aamevw fs maybe you have a mc of (01:10:05)
- ubiquiwus Nemßrknccess So thatyou can gen warme Hamm mmugn smndard mechanssmsthrough hecefoge eou mm mw c en s 1 5 very avaw ame fu Wu (01:10:05)
- a age e e r se e e e ouanemaun SeIfrService This is an about me bus ness user ng me cm sbbb but me busmess user bemgame tn decide I want mis and ggf get ir themselves un latera v gmbs rc an me c o shop maybe managingic or overseeing im ensure security and cgmb iance ana sa farm But fr enables me business user tu getaccess rg cesbrrg agv (01:10:05)
- cum Computing 5 3 modek for enam ng avai ab e mnvenien nnfdemand ewmfk access m a sharedpu of cunfigurab e compuüng fesuumesmaman be raur y Dmvxsmea am reieased Wim mmxmm FM f WM (01:10:05)
- Jokes With Clouds (01:11:17)
- Jokes with Clouds (01:11:45)
- As ven wazewsu zzb nm ou zv wcessss weze FAwo coMP s (01:11:45)
- Infrastructure (01:12:42)
- Breaclth (01:12:42)
- Platform (01:12:42)
- Niche (01:12:42)
- IaaS PaaS SaaS C OLlCÜ C tiOl 1 (01:14:36)
- Service Value to Users (01:19:14)
- Why will the cloucl always win (over loc l computing)? (01:20:56)
- Overview (01:23:24)
- SaaS Software as a Service (01:23:45)
- Maciern Distributed IT Architecture (01:25:30)
- ATTYBZOHI OWHS Ü ta Cel tE S SIOFEQS B d COfT1 lJC8 services (01:25:30)
- Sü GSf0I CEZ REHTS Ü tö CEHCEFS provides CRM 5O fW Fe (01:25:30)
- Hardware Mass ve geograph a c stnr ute farrns of commodnzy components (01:25:30)
- Two Types of Costs (01:25:33)
- Two Cost Recluction Techniques (01:27:08)
- Features versus Costs (01:32:39)


























































































