Block Party Application Packet: Fill & Download for Free

The DAAD webpage is an excellent starting point for anybody who wants to pursue their Master's in Germany (link below)Deutscher Akademischer Austausch DienstThe typical procedure for application would look like the one below..Pre-Application phase : If you are looking for a English taught program, then most of the Universities require you to produce a IELTS/TOEFL score card as a test of your English language proficiency. A score of 6.0+ in IELTS and 90+ in TOEFL would be sufficient enough to meet the requirements from most of the universities. Another point to be noted is that IELTS/TOEFL score card is mandatory for the German student Visa application. GRE score report is not mandatory for all the universities, while some of the reputed courses in universities like TU Munich, RWTH Aachen do ask for the same.Application phase : Most of the universities would expect an application packet to contain the following documents. a) Statement of Purpose / Motivation letter. b) Curriculum Vitae (Europass Format) c) University Transcript (Attested copy from the university) d) Letter of recommendation ( from your college Professors - highly preferred for application to the TU's, from your workplace - preferably from a person in Lead/Managerial position ) e) Attested copy of German Language certificates (if specified by the university ) f) Attested copy of IELTS/TOEFL score report. g) Work Experience letter (if any) h) The duly filled and signed application itself !!! Once all the above documents are ready, you have to send it to the address specified by the university through post (use DHL coupons if you are applying for 4+ universities ). However, some universities accept applications through a 3rd party called Uni-Assist in which case, you got to fill the application online in the Uni-Assist portal, attach the relevant documents and post it to the address specified by the Uni-Assist web portal. Make sure you send in your application much before the deadline in order to have sufficient breathing time for other activities that follow.Post-Application phase [Waiting for Feedback] : Once the application phase is complete, you need to wait for some weeks to receive the status of your application. Some universities like TU Munich, TU Darmstadt provide a status update of your application via their application portal, while some inform you by mail. The average waiting time to receive a Admit/Reject information is around 4 weeks, while it may vary between 3-8 weeks between different universities.Accepting the Admit + Visa process : After accepting the admit, you need to open a blocked account and transfer a sum of 8080 Euros (some consulates only accept the Visa application if the amount is transferred) for the Visa process. a) The Blocked account form can be found here - Page on www.deutsche-bank.de. You need to fill the form, take it to the German Consulate in your city along with your passport and get a attestation stamp on the form ( which costs 20 Euros ) from them and post it to the address mentioned on the form. You get a mail from the bank where you've opened the blocked account containing information like the IBAN number etc. within a week or two. b) The next step is to book a Visa appointment at your nearest German Consulate and follow the information provided by them w.r.t the documents that need to be carried for the Visa interview.Looking out for accommodation : New country, new people, new language and then finding accommodation is really a tedious task for anyone. Some universities provide accommodation, while some dont. The best way is to contact anyone through Facebook groups to understand the best possible way to find accommodation in a particular city. The following link might also prove to be helpful : Your flat share, apartment, housing search specialists for landlords and tenants: WG-Gesucht.deAll the very best for your application process !! :)

There can be many possibilities why the connection is getting refused.Firewall configuration- The connection might be refused either by the host firewall or a gateway firewall. In any case, you should change the firewall settings to allow out comming packets. You should also check if the OS has blocked the port in which the service is running. In the gateway, you should consider enlisting the port number and disable any packet filters.SSH server- it might be possible that your SSH config file doesn’t allow a third-party service such as VNC to establish a SSH connection. Otherwise, you would have to update your SSL certificate.Port already in use- another possibility is that the port used for VNC sessions may be used by some other third-party application. In that case, you need to stop that application/process(.exe) using task manager or migrate that application to another port, by changing the .config file.

Let's evaluate the problem a bit better.We are trying to send a BIG array of doubles using UDP. For manyapplications, the receiving party could only do something useful if it:Has the entire array buffered in memory, so that it can iterate over it.Receives the elements of the array in order. It probably should receive [0:10] before it can handle [11:20] etc.If both of these are true, then the TCP is probably the way to go. It already provides facilities for reliable in-order delivery.That being said, UDP could be a fine choice if the receiving party is doing some calculation where data is streamed and order of receiving elements does not matter. Further if the receiving party does not have enough memory to the entire array, receiving small chunks of it at a time is a useful strategy. A few examples of such applications is calculating the average or median of the array. These kinds of applications do not need the order guarantee of TCP but probably still need reliable delivery. So whatever protocol we design needs to account for packet loss. A few techniques that you can look up to see how such protocols are designed are:ACK based techniques vs NACK/NAK based techniques. You could also have a mix of both techniques.Sliding window protocols are useful in reducing the ping-pong nature of most ACK protocols.Sending parity packets. The idea is that with every n packets you send an additional packet such that if any one of the n + 1 (with the redundant packet) packets are lost, you can recover the information from the remaining n packets. A simple technique is to XOR the contents of n packets into the (n+1)th packet. Choosing the right value of n is a skill that among other things depends on the reliability of the network.So a high level solution requires you to break the double array into smaller chunks such that loss of any single packet does not cause a big re-transmission cycle. Again the ideal size of a chunk depends on many factors including the network conditions. You also want some framing on top of the actual array elements. The framing will serve multiple purposes including identifying duplicate data (you might not want to process data that you already have), identifying parity packets and recovering lost packets.As for the encoding of the data itself, it depends on the shape of your data. Is your data strictly non-negative? What's the variance? Is it sorted? Can you accept loss of precision? All of these factors can be used to come up with an ideal compression technique. Here are two links that I found useful : How to serialise an array of doubles with a byte (binary delta encoding for low-varianced monotonic sets of floating point data) | Java Code Geeks and Compression algorithm for IEEE-754 data. Any compression technique that you use should limit the number of packets across which you apply your compression. This will limit the number of packets the receiver has to buffer before it can decode the data and do something useful with it. For example: Let's assume your data requires limited precision, has low variance and is monotonically increasing. Converting to 64 bit integers and performing delta encoding on them might be a good format for transmission. If we choose to do delta encoding over the entire array then, losing the first packet means we cannot decode the contents of any of the latter packets till the 1st packet is recovered. So there is a trade-off between delta encoding too many packets (packet loss causes head of line blocking) and too few packets (more number of packets required to transmit the data).As for language specific techniques, in C++ encoding the data on top of a byte (unit8, char whatever) array is probably the way to go. That way you don't have an expensive serialize step that takes your double array and converts it into the compressed form. To do even less work at the time of serialization, your wrapper class could also write the framing information inline (into the byte array) with the actual data. Then you can just cast appropriate segments of the array and send them over the network. To do something similar in Java, you should probably use a ByteBuffer and create a wrapper class around it that does compression on the fly. For example if you had no need for compression, you could get a DoubleBuffer (Java Platform SE 7 ) from a ByteBuffer (Java Platform SE 7 ) and apply your operations on top of it. When sending the UDP packets out you can create slices from your ByteBuffer and send them out. Again if you wanted to inline the framing data then you'd have to create your own wrapper on top of a ByteBuffer. In Java it is especially useful to back your data with a ByteBuffer, otherwise network operations incur a copy.I wanted to show that efficiency of the format has more to it than meets the eye. The most compact encoding (for an in memory application) is not enough to get an efficient transfer over the network. One must account for the nature of UDP, the network itself, avoiding a serialization step etc when coming up with the "most efficient" format. Hope that gives you enough material to do more research and come up with a concrete solution.